6.6.66-rc1 review

Reviewed-by Mika Westerberg <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/watchdog/iTCO_wdt.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

drivers/watchdog/iTCO_wdt.c +19 -2

index 264857d314da8..dd297dcd524c9 100644
@@ -82,6 +82,13 @@
 #define TCO2_CNT(p)	(TCOBASE(p) + 0x0a) /* TCO2 Control Register	*/
 #define TCOv2_TMR(p)	(TCOBASE(p) + 0x12) /* TCOv2 Timer Initial Value*/
 
+/*
+ * NMI_NOW is bit 8 of TCO1_CNT register
+ * Read/Write
+ * This bit is implemented as RW but has no effect on HW.
+ */
+#define NMI_NOW		BIT(8)
+
 /* internal variables */
 struct iTCO_wdt_private {
 	struct watchdog_device wddev;

@@ -219,13 +226,23 @@ static int update_no_reboot_bit_cnt(void *priv, bool set)
 	struct iTCO_wdt_private *p = priv;
 	u16 val, newval;
 
-	val = inw(TCO1_CNT(p));
+	/*
+	 * writing back 1b1 to NMI_NOW of TCO1_CNT register
+	 * causes NMI_NOW bit inversion what consequently does
+	 * not allow to perform the register's value comparison
+	 * properly.
+	 *
+	 * NMI_NOW bit masking for TCO1_CNT register values
+	 * helps to avoid possible NMI_NOW bit inversions on
+	 * following write operation.
+	 */
+	val = inw(TCO1_CNT(p)) & ~NMI_NOW;
 	if (set)
 		val |= BIT(0);
 	else
 		val &= ~BIT(0);
 	outw(val, TCO1_CNT(p));
-	newval = inw(TCO1_CNT(p));
+	newval = inw(TCO1_CNT(p)) & ~NMI_NOW;
 
 	/* make sure the update is successful */
 	return val != newval ? -EIO : 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #2 lore raw

PATCH 2/356 watchdog: xilinx_wwdt: Calculate max_hw_heartbeat_ms using clock frequency

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Harini T <[email protected]>

[ Upstream commit 006778844c2c132c28cfa90e3570560351e01b9a ]

In the current implementation, the value of max_hw_heartbeat_ms is set to the timeout period expressed in milliseconds and fails to verify if the close window percentage exceeds the maximum value that the hardware supports.

1. Calculate max_hw_heartbeat_ms based on input clock frequency.
2. Update frequency check to require a minimum frequency of 1Mhz.
3. Limit the close and open window percent to hardware supported value to avoid truncation.
4. If the user input timeout exceeds the maximum timeout supported, use only open window and the framework supports the higher timeouts.

Fixes 12984cea1b8c ("watchdog: xilinx_wwdt: Add Versal window watchdog support")

Signed-off-by Harini T <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/watchdog/xilinx_wwdt.c | 75 ++++++++++++++++++++++++++++------
 1 file changed, 63 insertions(+), 12 deletions(-)

drivers/watchdog/xilinx_wwdt.c +63 ~~-12~~

index d271e2e8d6e27..3d2a156f71800 100644
@@ -2,7 +2,7 @@
 /*
  * Window watchdog device driver for Xilinx Versal WWDT
  *
- * Copyright (C) 2022 - 2023, Advanced Micro Devices, Inc.
+ * Copyright (C) 2022 - 2024, Advanced Micro Devices, Inc.
  */
 
 #include <linux/clk.h>

@@ -36,6 +36,12 @@
 
 #define XWWDT_CLOSE_WINDOW_PERCENT	50
 
+/* Maximum count value of each 32 bit window */
+#define XWWDT_MAX_COUNT_WINDOW		GENMASK(31, 0)
+
+/* Maximum count value of closed and open window combined */
+#define XWWDT_MAX_COUNT_WINDOW_COMBINED	GENMASK_ULL(32, 1)
+
 static int wwdt_timeout;
 static int closed_window_percent;

@@ -54,6 +60,8 @@ MODULE_PARM_DESC(closed_window_percent,
  * @xilinx_wwdt_wdd: watchdog device structure
  * @freq: source clock frequency of WWDT
  * @close_percent: Closed window percent
+ * @closed_timeout: Closed window timeout in ticks
+ * @open_timeout: Open window timeout in ticks
  */
 struct xwwdt_device {
 	void __iomem *base;

@@ -61,27 +69,22 @@ struct xwwdt_device {
 	struct watchdog_device xilinx_wwdt_wdd;
 	unsigned long freq;
 	u32 close_percent;
+	u64 closed_timeout;
+	u64 open_timeout;
 };
 
 static int xilinx_wwdt_start(struct watchdog_device *wdd)
 {
 	struct xwwdt_device *xdev = watchdog_get_drvdata(wdd);
 	struct watchdog_device *xilinx_wwdt_wdd = &xdev->xilinx_wwdt_wdd;
-	u64 time_out, closed_timeout, open_timeout;
 	u32 control_status_reg;
 
-	/* Calculate timeout count */
-	time_out = xdev->freq * wdd->timeout;
-	closed_timeout = div_u64(time_out * xdev->close_percent, 100);
-	open_timeout = time_out - closed_timeout;
-	wdd->min_hw_heartbeat_ms = xdev->close_percent * 10 * wdd->timeout;
-
 	spin_lock(&xdev->spinlock);
 
 	iowrite32(XWWDT_MWR_MASK, xdev->base + XWWDT_MWR_OFFSET);
 	iowrite32(~(u32)XWWDT_ESR_WEN_MASK, xdev->base + XWWDT_ESR_OFFSET);
-	iowrite32((u32)closed_timeout, xdev->base + XWWDT_FWR_OFFSET);
-	iowrite32((u32)open_timeout, xdev->base + XWWDT_SWR_OFFSET);
+	iowrite32((u32)xdev->closed_timeout, xdev->base + XWWDT_FWR_OFFSET);
+	iowrite32((u32)xdev->open_timeout, xdev->base + XWWDT_SWR_OFFSET);
 
 	/* Enable the window watchdog timer */
 	control_status_reg = ioread32(xdev->base + XWWDT_ESR_OFFSET);

@@ -133,7 +136,12 @@ static int xwwdt_probe(struct platform_device *pdev)
 	struct watchdog_device *xilinx_wwdt_wdd;
 	struct device *dev = &pdev->dev;
 	struct xwwdt_device *xdev;
+	u64 max_per_window_ms;
+	u64 min_per_window_ms;
+	u64 timeout_count;
 	struct clk *clk;
+	u32 timeout_ms;
+	u64 ms_count;
 	int ret;
 
 	xdev = devm_kzalloc(dev, sizeof(*xdev), GFP_KERNEL);

@@ -154,12 +162,13 @@ static int xwwdt_probe(struct platform_device *pdev)
 		return PTR_ERR(clk);
 
 	xdev->freq = clk_get_rate(clk);
-	if (!xdev->freq)
+	if (xdev->freq < 1000000)
 		return -EINVAL;
 
 	xilinx_wwdt_wdd->min_timeout = XWWDT_MIN_TIMEOUT;
 	xilinx_wwdt_wdd->timeout = XWWDT_DEFAULT_TIMEOUT;
-	xilinx_wwdt_wdd->max_hw_heartbeat_ms = 1000 * xilinx_wwdt_wdd->timeout;
+	xilinx_wwdt_wdd->max_hw_heartbeat_ms =
+		div64_u64(XWWDT_MAX_COUNT_WINDOW_COMBINED, xdev->freq) * 1000;
 
 	if (closed_window_percent == 0 || closed_window_percent >= 100)
 		xdev->close_percent = XWWDT_CLOSE_WINDOW_PERCENT;

@@ -167,6 +176,48 @@ static int xwwdt_probe(struct platform_device *pdev)
 		xdev->close_percent = closed_window_percent;
 
 	watchdog_init_timeout(xilinx_wwdt_wdd, wwdt_timeout, &pdev->dev);
+
+	/* Calculate ticks for 1 milli-second */
+	ms_count = div_u64(xdev->freq, 1000);
+	timeout_ms = xilinx_wwdt_wdd->timeout * 1000;
+	timeout_count = timeout_ms * ms_count;
+
+	if (timeout_ms > xilinx_wwdt_wdd->max_hw_heartbeat_ms) {
+		/*
+		 * To avoid ping restrictions until the minimum hardware heartbeat,
+		 * we will solely rely on the open window and
+		 * adjust the minimum hardware heartbeat to 0.
+		 */
+		xdev->closed_timeout = 0;
+		xdev->open_timeout = XWWDT_MAX_COUNT_WINDOW;
+		xilinx_wwdt_wdd->min_hw_heartbeat_ms = 0;
+		xilinx_wwdt_wdd->max_hw_heartbeat_ms = xilinx_wwdt_wdd->max_hw_heartbeat_ms / 2;
+	} else {
+		xdev->closed_timeout  = div64_u64(timeout_count * xdev->close_percent, 100);
+		xilinx_wwdt_wdd->min_hw_heartbeat_ms =
+			div64_u64(timeout_ms * xdev->close_percent, 100);
+
+		if (timeout_ms > xilinx_wwdt_wdd->max_hw_heartbeat_ms / 2) {
+			max_per_window_ms = xilinx_wwdt_wdd->max_hw_heartbeat_ms / 2;
+			min_per_window_ms = timeout_ms - max_per_window_ms;
+
+			if (xilinx_wwdt_wdd->min_hw_heartbeat_ms > max_per_window_ms) {
+				dev_info(xilinx_wwdt_wdd->parent,
+					 "Closed window cannot be set to %d%%. Using maximum supported value.\n",
+					xdev->close_percent);
+				xdev->closed_timeout = max_per_window_ms * ms_count;
+				xilinx_wwdt_wdd->min_hw_heartbeat_ms = max_per_window_ms;
+			} else if (xilinx_wwdt_wdd->min_hw_heartbeat_ms < min_per_window_ms) {
+				dev_info(xilinx_wwdt_wdd->parent,
+					 "Closed window cannot be set to %d%%. Using minimum supported value.\n",
+					xdev->close_percent);
+				xdev->closed_timeout = min_per_window_ms * ms_count;
+				xilinx_wwdt_wdd->min_hw_heartbeat_ms = min_per_window_ms;
+			}
+		}
+		xdev->open_timeout = timeout_count - xdev->closed_timeout;
+	}
+
 	spin_lock_init(&xdev->spinlock);
 	watchdog_set_drvdata(xilinx_wwdt_wdd, xdev);
 	watchdog_set_nowayout(xilinx_wwdt_wdd, 1);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #3 lore raw

PATCH 3/356 watchdog: apple: Actually flush writes after requesting watchdog restart

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nick Chan <[email protected]>

[ Upstream commit 51dfe714c03c066aabc815a2bb2adcc998dfcb30 ]

Although there is an existing code comment about flushing the writes, writes were not actually being flushed.

Actually flush the writes by changing readl_relaxed() to readl().

Fixes 4ed224aeaf661 ("watchdog: Add Apple SoC watchdog driver")

Signed-off-by Nick Chan <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/watchdog/apple_wdt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/watchdog/apple_wdt.c +1 -1

index eddeb0fede896..24e457b695662 100644
@@ -130,7 +130,7 @@ static int apple_wdt_restart(struct watchdog_device *wdd, unsigned long mode,
 	 * can take up to ~20-25ms until the SoC is actually reset. Just wait
 	 * 50ms here to be safe.
 	 */
-	(void)readl_relaxed(wdt->regs + APPLE_WDT_WD1_CUR_TIME);
+	(void)readl(wdt->regs + APPLE_WDT_WD1_CUR_TIME);
 	mdelay(50);
 
 	return 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #4 lore raw

PATCH 4/356 watchdog: mediatek: Make sure system reset gets asserted in mtk_wdt_restart()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yassine Oudjana <[email protected]>

[ Upstream commit a1495a21e0b8aad92132dfcf9c6fffc1bde9d5b2 ]

Clear the IRQ enable bit of WDT_MODE before asserting software reset in order to make TOPRGU issue a system reset signal instead of an IRQ.

Fixes a44a45536f7b ("watchdog: Add driver for Mediatek watchdog")

Signed-off-by Yassine Oudjana <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/watchdog/mtk_wdt.c | 6 ++++++
 1 file changed, 6 insertions(+)

drivers/watchdog/mtk_wdt.c +6 -0

index b2330b16b497a..0559d9f2d97b3 100644
@@ -186,9 +186,15 @@ static int mtk_wdt_restart(struct watchdog_device *wdt_dev,
 {
 	struct mtk_wdt_dev *mtk_wdt = watchdog_get_drvdata(wdt_dev);
 	void __iomem *wdt_base;
+	u32 reg;
 
 	wdt_base = mtk_wdt->wdt_base;
 
+	/* Enable reset in order to issue a system reset instead of an IRQ */
+	reg = readl(wdt_base + WDT_MODE);
+	reg &= ~WDT_MODE_IRQ_EN;
+	writel(reg | WDT_MODE_KEY, wdt_base + WDT_MODE);
+
 	while (1) {
 		writel(WDT_SWRST_KEY, wdt_base + WDT_SWRST);
 		mdelay(5);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #5 lore raw

PATCH 5/356 can: gs_usb: add VID/PID for Xylanta SAINT3 product family

Link https://lore.kernel.org/all/[email protected] open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marc Kleine-Budde <[email protected]>

[ Upstream commit 69e2326a21ef409d6c709cb990565331727b9f27 ]

Add support for the Xylanta SAINT3 product family.

Cc Andy Jackson <[email protected]>

Cc Ken Aitchison <[email protected]>

Tested-by Andy Jackson <[email protected]>

Signed-off-by Marc Kleine-Budde <[email protected]>

Stable-dep-of: 889b2ae9139a ("can: gs_usb: add usb endpoint address detection at driver probe step")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/usb/gs_usb.c | 5 +++++
 1 file changed, 5 insertions(+)

drivers/net/can/usb/gs_usb.c +5 -0

index 95b0fdb602c8f..ed59451f03cdd 100644
@@ -40,6 +40,9 @@
 #define USB_ABE_CANDEBUGGER_FD_VENDOR_ID 0x16d0
 #define USB_ABE_CANDEBUGGER_FD_PRODUCT_ID 0x10b8
 
+#define USB_XYLANTA_SAINT3_VENDOR_ID 0x16d0
+#define USB_XYLANTA_SAINT3_PRODUCT_ID 0x0f30
+
 #define GS_USB_ENDPOINT_IN 1
 #define GS_USB_ENDPOINT_OUT 2

@@ -1530,6 +1533,8 @@ static const struct usb_device_id gs_usb_table[] = {
 				      USB_CES_CANEXT_FD_PRODUCT_ID, 0) },
 	{ USB_DEVICE_INTERFACE_NUMBER(USB_ABE_CANDEBUGGER_FD_VENDOR_ID,
 				      USB_ABE_CANDEBUGGER_FD_PRODUCT_ID, 0) },
+	{ USB_DEVICE_INTERFACE_NUMBER(USB_XYLANTA_SAINT3_VENDOR_ID,
+				      USB_XYLANTA_SAINT3_PRODUCT_ID, 0) },
 	{} /* Terminating entry */
 };

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #6 lore raw

PATCH 6/356 can: gs_usb: add usb endpoint address detection at driver probe step

https://github.com/zephyrproject-rtos/zephyr/issues/67812

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexander Kozhinov <[email protected]>

[ Upstream commit 889b2ae9139a87b3390f7003cb1bb3d65bf90a26 ]

There is an approach made to implement gs_usb firmware/driver based on Zephyr RTOS. It was found that USB stack of Zephyr RTOS overwrites USB EP addresses, if they have different last 4 bytes in absence of other endpoints.

For example in case of gs_usb candlelight firmware EP-IN is 0x81 and EP-OUT 0x02. If there are no additional USB endpoints, Zephyr RTOS will overwrite EP-OUT to 0x01. More information can be found in the discussion with Zephyr RTOS USB stack maintainer here:

There are already two different gs_usb FW driver implementations based on Zephyr RTOS:

1. https://github.com/CANnectivity/cannectivity
(by: https://github.com/henrikbrixandersen)
2. https://github.com/zephyrproject-rtos/zephyr/compare/main...KozhinovAlexander:zephyr:gs_usb
(by: https://github.com/KozhinovAlexander)

At the moment both Zephyr RTOS implementations use dummy USB endpoint, to overcome described USB stack behavior from Zephyr itself. Since Zephyr RTOS is intended to be used on microcontrollers with very constrained amount of resources (ROM, RAM) and additional endpoint requires memory, it is more convenient to update the gs_usb driver in the Linux kernel.

To fix this problem, update the gs_usb driver from using hard coded endpoint numbers to evaluate the endpoint descriptors and use the endpoints provided there.

Fixes d08e973a77d1 ("can: gs_usb: Added support for the GS_USB CAN devices")

Reviewed-by Vincent Mailhol <[email protected]>

Signed-off-by Alexander Kozhinov <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/usb/gs_usb.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

drivers/net/can/usb/gs_usb.c +18 -7

index ed59451f03cdd..de616d6589c0b 100644
@@ -43,9 +43,6 @@
 #define USB_XYLANTA_SAINT3_VENDOR_ID 0x16d0
 #define USB_XYLANTA_SAINT3_PRODUCT_ID 0x0f30
 
-#define GS_USB_ENDPOINT_IN 1
-#define GS_USB_ENDPOINT_OUT 2
-
 /* Timestamp 32 bit timer runs at 1 MHz (1 µs tick). Worker accounts
  * for timer overflow (will be after ~71 minutes)
  */

@@ -336,6 +333,9 @@ struct gs_usb {
 
 	unsigned int hf_size_rx;
 	u8 active_channels;
+
+	unsigned int pipe_in;
+	unsigned int pipe_out;
 };
 
 /* 'allocate' a tx context.

@@ -687,7 +687,7 @@ static void gs_usb_receive_bulk_callback(struct urb *urb)
 
 resubmit_urb:
 	usb_fill_bulk_urb(urb, parent->udev,
-			  usb_rcvbulkpipe(parent->udev, GS_USB_ENDPOINT_IN),
+			  parent->pipe_in,
 			  hf, dev->parent->hf_size_rx,
 			  gs_usb_receive_bulk_callback, parent);

@@ -819,7 +819,7 @@ static netdev_tx_t gs_can_start_xmit(struct sk_buff *skb,
 	}
 
 	usb_fill_bulk_urb(urb, dev->udev,
-			  usb_sndbulkpipe(dev->udev, GS_USB_ENDPOINT_OUT),
+			  dev->parent->pipe_out,
 			  hf, dev->hf_size_tx,
 			  gs_usb_xmit_callback, txc);

@@ -925,8 +925,7 @@ static int gs_can_open(struct net_device *netdev)
 			/* fill, anchor, and submit rx urb */
 			usb_fill_bulk_urb(urb,
 					  dev->udev,
-					  usb_rcvbulkpipe(dev->udev,
-							  GS_USB_ENDPOINT_IN),
+					  dev->parent->pipe_in,
 					  buf,
 					  dev->parent->hf_size_rx,
 					  gs_usb_receive_bulk_callback, parent);

@@ -1413,6 +1412,7 @@ static int gs_usb_probe(struct usb_interface *intf,
 			const struct usb_device_id *id)
 {
 	struct usb_device *udev = interface_to_usbdev(intf);
+	struct usb_endpoint_descriptor *ep_in, *ep_out;
 	struct gs_host_frame *hf;
 	struct gs_usb *parent;
 	struct gs_host_config hconf = {

@@ -1422,6 +1422,13 @@ static int gs_usb_probe(struct usb_interface *intf,
 	unsigned int icount, i;
 	int rc;
 
+	rc = usb_find_common_endpoints(intf->cur_altsetting,
+				       &ep_in, &ep_out, NULL, NULL);
+	if (rc) {
+		dev_err(&intf->dev, "Required endpoints not found\n");
+		return rc;
+	}
+
 	/* send host config */
 	rc = usb_control_msg_send(udev, 0,
 				  GS_USB_BREQ_HOST_FORMAT,

@@ -1466,6 +1473,10 @@ static int gs_usb_probe(struct usb_interface *intf,
 	usb_set_intfdata(intf, parent);
 	parent->udev = udev;
 
+	/* store the detected endpoints */
+	parent->pipe_in = usb_rcvbulkpipe(parent->udev, ep_in->bEndpointAddress);
+	parent->pipe_out = usb_sndbulkpipe(parent->udev, ep_out->bEndpointAddress);
+
 	for (i = 0; i < icount; i++) {
 		unsigned int hf_size_rx = 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #7 lore raw

PATCH 7/356 can: c_can: c_can_handle_bus_err(): update statistics if skb allocation fails

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit 9e66242504f49e17481d8e197730faba7d99c934 ]

Ensure that the statistics are always updated, even if the skb allocation fails.

Fixes 4d6d26537940 ("can: c_can: fix {rx,tx}_errors statistics")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/c_can/c_can_main.c | 26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

drivers/net/can/c_can/c_can_main.c +17 -9

index 511615dc33419..cc371d0c9f3c7 100644
@@ -1014,49 +1014,57 @@ static int c_can_handle_bus_err(struct net_device *dev,
 
 	/* propagate the error condition to the CAN stack */
 	skb = alloc_can_err_skb(dev, &cf);
-	if (unlikely(!skb))
-		return 0;
 
 	/* check for 'last error code' which tells us the
 	 * type of the last error to occur on the CAN bus
 	 */
-	cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
+	if (likely(skb))
+		cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
 
 	switch (lec_type) {
 	case LEC_STUFF_ERROR:
 		netdev_dbg(dev, "stuff error\n");
-		cf->data[2] |= CAN_ERR_PROT_STUFF;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_STUFF;
 		stats->rx_errors++;
 		break;
 	case LEC_FORM_ERROR:
 		netdev_dbg(dev, "form error\n");
-		cf->data[2] |= CAN_ERR_PROT_FORM;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_FORM;
 		stats->rx_errors++;
 		break;
 	case LEC_ACK_ERROR:
 		netdev_dbg(dev, "ack error\n");
-		cf->data[3] = CAN_ERR_PROT_LOC_ACK;
+		if (likely(skb))
+			cf->data[3] = CAN_ERR_PROT_LOC_ACK;
 		stats->tx_errors++;
 		break;
 	case LEC_BIT1_ERROR:
 		netdev_dbg(dev, "bit1 error\n");
-		cf->data[2] |= CAN_ERR_PROT_BIT1;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_BIT1;
 		stats->tx_errors++;
 		break;
 	case LEC_BIT0_ERROR:
 		netdev_dbg(dev, "bit0 error\n");
-		cf->data[2] |= CAN_ERR_PROT_BIT0;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_BIT0;
 		stats->tx_errors++;
 		break;
 	case LEC_CRC_ERROR:
 		netdev_dbg(dev, "CRC error\n");
-		cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ;
+		if (likely(skb))
+			cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ;
 		stats->rx_errors++;
 		break;
 	default:
 		break;
 	}
 
+	if (unlikely(!skb))
+		return 0;
+
 	netif_receive_skb(skb);
 	return 1;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #8 lore raw

PATCH 8/356 can: sun4i_can: sun4i_can_err(): call can_change_state() even if cf is NULL

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit ee6bf3677ae03569d833795064e17f605c2163c7 ]

Call the function can_change_state() if the allocation of the skb fails, as it handles the cf parameter when it is null.

Additionally, this ensures that the statistics related to state error counters (i. e. warning, passive, and bus-off) are updated.

Fixes 0738eff14d81 ("can: Allwinner A10/A20 CAN Controller support - Kernel module")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/sun4i_can.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

drivers/net/can/sun4i_can.c +4 -4

index ab8d017846869..80f7aa2531dc0 100644
@@ -629,10 +629,10 @@ static int sun4i_can_err(struct net_device *dev, u8 isrc, u8 status)
 		tx_state = txerr >= rxerr ? state : 0;
 		rx_state = txerr <= rxerr ? state : 0;
 
-		if (likely(skb))
-			can_change_state(dev, cf, tx_state, rx_state);
-		else
-			priv->can.state = state;
+		/* The skb allocation might fail, but can_change_state()
+		 * handles cf == NULL.
+		 */
+		can_change_state(dev, cf, tx_state, rx_state);
 		if (state == CAN_STATE_BUS_OFF)
 			can_bus_off(dev);
 	}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #9 lore raw

PATCH 9/356 can: hi311x: hi3110_can_ist(): fix potential use-after-free

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit 9ad86d377ef4a19c75a9c639964879a5b25a433b ]

The commit a22bd630cfff ("can: hi311x: do not report txerr and rxerr during bus-off") removed the reporting of rxerr and txerr even in case of correct operation (i. e. not bus-off).

The error count information added to the CAN frame after netif_rx() is a potential use after free, since there is no guarantee that the skb is in the same state. It might be freed or reused.

Fix the issue by postponing the netif_rx() call in case of txerr and rxerr reporting.

Fixes a22bd630cfff ("can: hi311x: do not report txerr and rxerr during bus-off")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/spi/hi311x.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/net/can/spi/hi311x.c +2 -1

index e1b8533a602e2..fb58e294f7b79 100644
@@ -671,9 +671,9 @@ static irqreturn_t hi3110_can_ist(int irq, void *dev_id)
 			tx_state = txerr >= rxerr ? new_state : 0;
 			rx_state = txerr <= rxerr ? new_state : 0;
 			can_change_state(net, cf, tx_state, rx_state);
-			netif_rx(skb);
 
 			if (new_state == CAN_STATE_BUS_OFF) {
+				netif_rx(skb);
 				can_bus_off(net);
 				if (priv->can.restart_ms == 0) {
 					priv->force_quit = 1;

@@ -684,6 +684,7 @@ static irqreturn_t hi3110_can_ist(int irq, void *dev_id)
 				cf->can_id |= CAN_ERR_CNT;
 				cf->data[6] = txerr;
 				cf->data[7] = rxerr;
+				netif_rx(skb);
 			}
 		}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #10 lore raw

PATCH 10/356 can: m_can: m_can_handle_lec_err(): fix {rx,tx}_errors statistics

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit 988d4222bf9039a875a3d48f2fe35c317831ff68 ]

The m_can_handle_lec_err() function was incorrectly incrementing only the receive error counter, even in cases of bit or acknowledgment errors that occur during transmission.

Fix the issue by incrementing the appropriate counter based on the type of error.

Fixes e0d1f4816f2a ("can: m_can: add Bosch M_CAN controller support")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/m_can/m_can.c | 33 +++++++++++++++++++++++----------
 1 file changed, 23 insertions(+), 10 deletions(-)

drivers/net/can/m_can/m_can.c +23 ~~-10~~

index ec6e740b03247..2a258986eed02 100644
@@ -636,47 +636,60 @@ static int m_can_handle_lec_err(struct net_device *dev,
 	u32 timestamp = 0;
 
 	cdev->can.can_stats.bus_error++;
-	stats->rx_errors++;
 
 	/* propagate the error condition to the CAN stack */
 	skb = alloc_can_err_skb(dev, &cf);
-	if (unlikely(!skb))
-		return 0;
 
 	/* check for 'last error code' which tells us the
 	 * type of the last error to occur on the CAN bus
 	 */
-	cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
+	if (likely(skb))
+		cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
 
 	switch (lec_type) {
 	case LEC_STUFF_ERROR:
 		netdev_dbg(dev, "stuff error\n");
-		cf->data[2] |= CAN_ERR_PROT_STUFF;
+		stats->rx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_STUFF;
 		break;
 	case LEC_FORM_ERROR:
 		netdev_dbg(dev, "form error\n");
-		cf->data[2] |= CAN_ERR_PROT_FORM;
+		stats->rx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_FORM;
 		break;
 	case LEC_ACK_ERROR:
 		netdev_dbg(dev, "ack error\n");
-		cf->data[3] = CAN_ERR_PROT_LOC_ACK;
+		stats->tx_errors++;
+		if (likely(skb))
+			cf->data[3] = CAN_ERR_PROT_LOC_ACK;
 		break;
 	case LEC_BIT1_ERROR:
 		netdev_dbg(dev, "bit1 error\n");
-		cf->data[2] |= CAN_ERR_PROT_BIT1;
+		stats->tx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_BIT1;
 		break;
 	case LEC_BIT0_ERROR:
 		netdev_dbg(dev, "bit0 error\n");
-		cf->data[2] |= CAN_ERR_PROT_BIT0;
+		stats->tx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_BIT0;
 		break;
 	case LEC_CRC_ERROR:
 		netdev_dbg(dev, "CRC error\n");
-		cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ;
+		stats->rx_errors++;
+		if (likely(skb))
+			cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ;
 		break;
 	default:
 		break;
 	}
 
+	if (unlikely(!skb))
+		return 0;
+
 	if (cdev->is_peripheral)
 		timestamp = m_can_get_timestamp(cdev);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #11 lore raw

PATCH 11/356 can: ifi_canfd: ifi_canfd_handle_lec_err(): fix {rx,tx}_errors statistics

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit bb03d568bb21b4afe7935d1943bcf68ddea3ea45 ]

The ifi_canfd_handle_lec_err() function was incorrectly incrementing only the receive error counter, even in cases of bit or acknowledgment errors that occur during transmission.

Fix the issue by incrementing the appropriate counter based on the type of error.

Fixes 5bbd655a8bd0 ("can: ifi: Add more detailed error reporting")

Reviewed-by Marek Vasut <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/ifi_canfd/ifi_canfd.c | 58 ++++++++++++++++++---------
 1 file changed, 40 insertions(+), 18 deletions(-)

drivers/net/can/ifi_canfd/ifi_canfd.c +40 ~~-18~~

index 72307297d75e4..5145a6a73d2d7 100644
@@ -390,36 +390,55 @@ static int ifi_canfd_handle_lec_err(struct net_device *ndev)
 		return 0;
 
 	priv->can.can_stats.bus_error++;
-	stats->rx_errors++;
 
 	/* Propagate the error condition to the CAN stack. */
 	skb = alloc_can_err_skb(ndev, &cf);
-	if (unlikely(!skb))
-		return 0;
 
 	/* Read the error counter register and check for new errors. */
-	cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
+	if (likely(skb))
+		cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
 
-	if (errctr & IFI_CANFD_ERROR_CTR_OVERLOAD_FIRST)
-		cf->data[2] |= CAN_ERR_PROT_OVERLOAD;
+	if (errctr & IFI_CANFD_ERROR_CTR_OVERLOAD_FIRST) {
+		stats->rx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_OVERLOAD;
+	}
 
-	if (errctr & IFI_CANFD_ERROR_CTR_ACK_ERROR_FIRST)
-		cf->data[3] = CAN_ERR_PROT_LOC_ACK;
+	if (errctr & IFI_CANFD_ERROR_CTR_ACK_ERROR_FIRST) {
+		stats->tx_errors++;
+		if (likely(skb))
+			cf->data[3] = CAN_ERR_PROT_LOC_ACK;
+	}
 
-	if (errctr & IFI_CANFD_ERROR_CTR_BIT0_ERROR_FIRST)
-		cf->data[2] |= CAN_ERR_PROT_BIT0;
+	if (errctr & IFI_CANFD_ERROR_CTR_BIT0_ERROR_FIRST) {
+		stats->tx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_BIT0;
+	}
 
-	if (errctr & IFI_CANFD_ERROR_CTR_BIT1_ERROR_FIRST)
-		cf->data[2] |= CAN_ERR_PROT_BIT1;
+	if (errctr & IFI_CANFD_ERROR_CTR_BIT1_ERROR_FIRST) {
+		stats->tx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_BIT1;
+	}
 
-	if (errctr & IFI_CANFD_ERROR_CTR_STUFF_ERROR_FIRST)
-		cf->data[2] |= CAN_ERR_PROT_STUFF;
+	if (errctr & IFI_CANFD_ERROR_CTR_STUFF_ERROR_FIRST) {
+		stats->rx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_STUFF;
+	}
 
-	if (errctr & IFI_CANFD_ERROR_CTR_CRC_ERROR_FIRST)
-		cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ;
+	if (errctr & IFI_CANFD_ERROR_CTR_CRC_ERROR_FIRST) {
+		stats->rx_errors++;
+		if (likely(skb))
+			cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ;
+	}
 
-	if (errctr & IFI_CANFD_ERROR_CTR_FORM_ERROR_FIRST)
-		cf->data[2] |= CAN_ERR_PROT_FORM;
+	if (errctr & IFI_CANFD_ERROR_CTR_FORM_ERROR_FIRST) {
+		stats->rx_errors++;
+		if (likely(skb))
+			cf->data[2] |= CAN_ERR_PROT_FORM;
+	}
 
 	/* Reset the error counter, ack the IRQ and re-enable the counter. */
 	writel(IFI_CANFD_ERROR_CTR_ER_RESET, priv->base + IFI_CANFD_ERROR_CTR);

@@ -427,6 +446,9 @@ static int ifi_canfd_handle_lec_err(struct net_device *ndev)
 	       priv->base + IFI_CANFD_INTERRUPT);
 	writel(IFI_CANFD_ERROR_CTR_ER_ENABLE, priv->base + IFI_CANFD_ERROR_CTR);
 
+	if (unlikely(!skb))
+		return 0;
+
 	netif_receive_skb(skb);
 
 	return 1;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #12 lore raw

PATCH 12/356 can: hi311x: hi3110_can_ist(): fix {rx,tx}_errors statistics

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit 3e4645931655776e757f9fb5ae29371cd7cb21a2 ]

The hi3110_can_ist() function was incorrectly incrementing only the receive error counter, even in cases of bit or acknowledgment errors that occur during transmission.

The fix the issue by incrementing the appropriate counter based on the type of error.

Fixes 57e83fb9b746 ("can: hi311x: Add Holt HI-311x CAN driver")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/spi/hi311x.c | 47 ++++++++++++++++++++++--------------
 1 file changed, 29 insertions(+), 18 deletions(-)

drivers/net/can/spi/hi311x.c +29 ~~-18~~

index fb58e294f7b79..b757555ed4c4f 100644
@@ -697,27 +697,38 @@ static irqreturn_t hi3110_can_ist(int irq, void *dev_id)
 			/* Check for protocol errors */
 			if (eflag & HI3110_ERR_PROTOCOL_MASK) {
 				skb = alloc_can_err_skb(net, &cf);
-				if (!skb)
-					break;
+				if (skb)
+					cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
 
-				cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
 				priv->can.can_stats.bus_error++;
-				priv->net->stats.rx_errors++;
-				if (eflag & HI3110_ERR_BITERR)
-					cf->data[2] |= CAN_ERR_PROT_BIT;
-				else if (eflag & HI3110_ERR_FRMERR)
-					cf->data[2] |= CAN_ERR_PROT_FORM;
-				else if (eflag & HI3110_ERR_STUFERR)
-					cf->data[2] |= CAN_ERR_PROT_STUFF;
-				else if (eflag & HI3110_ERR_CRCERR)
-					cf->data[3] |= CAN_ERR_PROT_LOC_CRC_SEQ;
-				else if (eflag & HI3110_ERR_ACKERR)
-					cf->data[3] |= CAN_ERR_PROT_LOC_ACK;
-
-				cf->data[6] = hi3110_read(spi, HI3110_READ_TEC);
-				cf->data[7] = hi3110_read(spi, HI3110_READ_REC);
+				if (eflag & HI3110_ERR_BITERR) {
+					priv->net->stats.tx_errors++;
+					if (skb)
+						cf->data[2] |= CAN_ERR_PROT_BIT;
+				} else if (eflag & HI3110_ERR_FRMERR) {
+					priv->net->stats.rx_errors++;
+					if (skb)
+						cf->data[2] |= CAN_ERR_PROT_FORM;
+				} else if (eflag & HI3110_ERR_STUFERR) {
+					priv->net->stats.rx_errors++;
+					if (skb)
+						cf->data[2] |= CAN_ERR_PROT_STUFF;
+				} else if (eflag & HI3110_ERR_CRCERR) {
+					priv->net->stats.rx_errors++;
+					if (skb)
+						cf->data[3] |= CAN_ERR_PROT_LOC_CRC_SEQ;
+				} else if (eflag & HI3110_ERR_ACKERR) {
+					priv->net->stats.tx_errors++;
+					if (skb)
+						cf->data[3] |= CAN_ERR_PROT_LOC_ACK;
+				}
+
 				netdev_dbg(priv->net, "Bus Error\n");
-				netif_rx(skb);
+				if (skb) {
+					cf->data[6] = hi3110_read(spi, HI3110_READ_TEC);
+					cf->data[7] = hi3110_read(spi, HI3110_READ_REC);
+					netif_rx(skb);
+				}
 			}
 		}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #13 lore raw

PATCH 13/356 can: sja1000: sja1000_err(): fix {rx,tx}_errors statistics

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit 2c4ef3af4b028a0eaaf378df511d3b425b1df61f ]

The sja1000_err() function only incremented the receive error counter and never the transmit error counter, even if the ECC_DIR flag reported that an error had occurred during transmission.

Increment the receive/transmit error counter based on the value of the ECC_DIR flag.

Fixes 429da1cc841b ("can: Driver for the SJA1000 CAN controller")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/sja1000/sja1000.c | 67 ++++++++++++++++++-------------
 1 file changed, 39 insertions(+), 28 deletions(-)

drivers/net/can/sja1000/sja1000.c +39 ~~-28~~

index 743c2eb62b877..10e211d917e31 100644
@@ -416,8 +416,6 @@ static int sja1000_err(struct net_device *dev, uint8_t isrc, uint8_t status)
 	int ret = 0;
 
 	skb = alloc_can_err_skb(dev, &cf);
-	if (skb == NULL)
-		return -ENOMEM;
 
 	txerr = priv->read_reg(priv, SJA1000_TXERR);
 	rxerr = priv->read_reg(priv, SJA1000_RXERR);

@@ -425,8 +423,11 @@ static int sja1000_err(struct net_device *dev, uint8_t isrc, uint8_t status)
 	if (isrc & IRQ_DOI) {
 		/* data overrun interrupt */
 		netdev_dbg(dev, "data overrun interrupt\n");
-		cf->can_id |= CAN_ERR_CRTL;
-		cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW;
+		if (skb) {
+			cf->can_id |= CAN_ERR_CRTL;
+			cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW;
+		}
+
 		stats->rx_over_errors++;
 		stats->rx_errors++;
 		sja1000_write_cmdreg(priv, CMD_CDO);	/* clear bit */

@@ -452,7 +453,7 @@ static int sja1000_err(struct net_device *dev, uint8_t isrc, uint8_t status)
 		else
 			state = CAN_STATE_ERROR_ACTIVE;
 	}
-	if (state != CAN_STATE_BUS_OFF) {
+	if (state != CAN_STATE_BUS_OFF && skb) {
 		cf->can_id |= CAN_ERR_CNT;
 		cf->data[6] = txerr;
 		cf->data[7] = rxerr;

@@ -460,33 +461,38 @@ static int sja1000_err(struct net_device *dev, uint8_t isrc, uint8_t status)
 	if (isrc & IRQ_BEI) {
 		/* bus error interrupt */
 		priv->can.can_stats.bus_error++;
-		stats->rx_errors++;
 
 		ecc = priv->read_reg(priv, SJA1000_ECC);
+		if (skb) {
+			cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
 
-		cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
-
-		/* set error type */
-		switch (ecc & ECC_MASK) {
-		case ECC_BIT:
-			cf->data[2] |= CAN_ERR_PROT_BIT;
-			break;
-		case ECC_FORM:
-			cf->data[2] |= CAN_ERR_PROT_FORM;
-			break;
-		case ECC_STUFF:
-			cf->data[2] |= CAN_ERR_PROT_STUFF;
-			break;
-		default:
-			break;
-		}
+			/* set error type */
+			switch (ecc & ECC_MASK) {
+			case ECC_BIT:
+				cf->data[2] |= CAN_ERR_PROT_BIT;
+				break;
+			case ECC_FORM:
+				cf->data[2] |= CAN_ERR_PROT_FORM;
+				break;
+			case ECC_STUFF:
+				cf->data[2] |= CAN_ERR_PROT_STUFF;
+				break;
+			default:
+				break;
+			}
 
-		/* set error location */
-		cf->data[3] = ecc & ECC_SEG;
+			/* set error location */
+			cf->data[3] = ecc & ECC_SEG;
+		}
 
 		/* Error occurred during transmission? */
-		if ((ecc & ECC_DIR) == 0)
-			cf->data[2] |= CAN_ERR_PROT_TX;
+		if ((ecc & ECC_DIR) == 0) {
+			stats->tx_errors++;
+			if (skb)
+				cf->data[2] |= CAN_ERR_PROT_TX;
+		} else {
+			stats->rx_errors++;
+		}
 	}
 	if (isrc & IRQ_EPI) {
 		/* error passive interrupt */

@@ -502,8 +508,10 @@ static int sja1000_err(struct net_device *dev, uint8_t isrc, uint8_t status)
 		netdev_dbg(dev, "arbitration lost interrupt\n");
 		alc = priv->read_reg(priv, SJA1000_ALC);
 		priv->can.can_stats.arbitration_lost++;
-		cf->can_id |= CAN_ERR_LOSTARB;
-		cf->data[0] = alc & 0x1f;
+		if (skb) {
+			cf->can_id |= CAN_ERR_LOSTARB;
+			cf->data[0] = alc & 0x1f;
+		}
 	}
 
 	if (state != priv->can.state) {

@@ -516,6 +524,9 @@ static int sja1000_err(struct net_device *dev, uint8_t isrc, uint8_t status)
 			can_bus_off(dev);
 	}
 
+	if (!skb)
+		return -ENOMEM;
+
 	netif_rx(skb);
 
 	return ret;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #14 lore raw

PATCH 14/356 can: sun4i_can: sun4i_can_err(): fix {rx,tx}_errors statistics

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit 595a81988a6fe06eb5849e972c8b9cb21c4e0d54 ]

The sun4i_can_err() function only incremented the receive error counter and never the transmit error counter, even if the STA_ERR_DIR flag reported that an error had occurred during transmission.

Increment the receive/transmit error counter based on the value of the STA_ERR_DIR flag.

Fixes 0738eff14d81 ("can: Allwinner A10/A20 CAN Controller support - Kernel module")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/sun4i_can.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

drivers/net/can/sun4i_can.c +9 -5

index 80f7aa2531dc0..283fbf59e66d5 100644
@@ -579,11 +579,9 @@ static int sun4i_can_err(struct net_device *dev, u8 isrc, u8 status)
 		/* bus error interrupt */
 		netdev_dbg(dev, "bus error interrupt\n");
 		priv->can.can_stats.bus_error++;
-		stats->rx_errors++;
+		ecc = readl(priv->base + SUN4I_REG_STA_ADDR);
 
 		if (likely(skb)) {
-			ecc = readl(priv->base + SUN4I_REG_STA_ADDR);
-
 			cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
 
 			switch (ecc & SUN4I_STA_MASK_ERR) {

@@ -601,9 +599,15 @@ static int sun4i_can_err(struct net_device *dev, u8 isrc, u8 status)
 					       >> 16;
 				break;
 			}
-			/* error occurred during transmission? */
-			if ((ecc & SUN4I_STA_ERR_DIR) == 0)
+		}
+
+		/* error occurred during transmission? */
+		if ((ecc & SUN4I_STA_ERR_DIR) == 0) {
+			if (likely(skb))
 				cf->data[2] |= CAN_ERR_PROT_TX;
+			stats->tx_errors++;
+		} else {
+			stats->rx_errors++;
 		}
 	}
 	if (isrc & SUN4I_INT_ERR_PASSIVE) {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #15 lore raw

PATCH 15/356 can: ems_usb: ems_usb_rx_err(): fix {rx,tx}_errors statistics

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit 72a7e2e74b3075959f05e622bae09b115957dffe ]

The ems_usb_rx_err() function only incremented the receive error counter and never the transmit error counter, even if the ECC_DIR flag reported that an error had occurred during transmission.

Increment the receive/transmit error counter based on the value of the ECC_DIR flag.

Fixes 702171adeed3 ("ems_usb: Added support for EMS CPC-USB/ARM7 CAN/USB interface")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/usb/ems_usb.c | 58 ++++++++++++++++++++---------------
 1 file changed, 33 insertions(+), 25 deletions(-)

drivers/net/can/usb/ems_usb.c +33 ~~-25~~

index 050c0b49938a4..5355bac4dccbe 100644
@@ -335,15 +335,14 @@ static void ems_usb_rx_err(struct ems_usb *dev, struct ems_cpc_msg *msg)
 	struct net_device_stats *stats = &dev->netdev->stats;
 
 	skb = alloc_can_err_skb(dev->netdev, &cf);
-	if (skb == NULL)
-		return;
 
 	if (msg->type == CPC_MSG_TYPE_CAN_STATE) {
 		u8 state = msg->msg.can_state;
 
 		if (state & SJA1000_SR_BS) {
 			dev->can.state = CAN_STATE_BUS_OFF;
-			cf->can_id |= CAN_ERR_BUSOFF;
+			if (skb)
+				cf->can_id |= CAN_ERR_BUSOFF;
 
 			dev->can.can_stats.bus_off++;
 			can_bus_off(dev->netdev);

@@ -361,44 +360,53 @@ static void ems_usb_rx_err(struct ems_usb *dev, struct ems_cpc_msg *msg)
 
 		/* bus error interrupt */
 		dev->can.can_stats.bus_error++;
-		stats->rx_errors++;
 
-		cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
+		if (skb) {
+			cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
 
-		switch (ecc & SJA1000_ECC_MASK) {
-		case SJA1000_ECC_BIT:
-			cf->data[2] |= CAN_ERR_PROT_BIT;
-			break;
-		case SJA1000_ECC_FORM:
-			cf->data[2] |= CAN_ERR_PROT_FORM;
-			break;
-		case SJA1000_ECC_STUFF:
-			cf->data[2] |= CAN_ERR_PROT_STUFF;
-			break;
-		default:
-			cf->data[3] = ecc & SJA1000_ECC_SEG;
-			break;
+			switch (ecc & SJA1000_ECC_MASK) {
+			case SJA1000_ECC_BIT:
+				cf->data[2] |= CAN_ERR_PROT_BIT;
+				break;
+			case SJA1000_ECC_FORM:
+				cf->data[2] |= CAN_ERR_PROT_FORM;
+				break;
+			case SJA1000_ECC_STUFF:
+				cf->data[2] |= CAN_ERR_PROT_STUFF;
+				break;
+			default:
+				cf->data[3] = ecc & SJA1000_ECC_SEG;
+				break;
+			}
 		}
 
 		/* Error occurred during transmission? */
-		if ((ecc & SJA1000_ECC_DIR) == 0)
-			cf->data[2] |= CAN_ERR_PROT_TX;
+		if ((ecc & SJA1000_ECC_DIR) == 0) {
+			stats->tx_errors++;
+			if (skb)
+				cf->data[2] |= CAN_ERR_PROT_TX;
+		} else {
+			stats->rx_errors++;
+		}
 
-		if (dev->can.state == CAN_STATE_ERROR_WARNING ||
-		    dev->can.state == CAN_STATE_ERROR_PASSIVE) {
+		if (skb && (dev->can.state == CAN_STATE_ERROR_WARNING ||
+			    dev->can.state == CAN_STATE_ERROR_PASSIVE)) {
 			cf->can_id |= CAN_ERR_CRTL;
 			cf->data[1] = (txerr > rxerr) ?
 			    CAN_ERR_CRTL_TX_PASSIVE : CAN_ERR_CRTL_RX_PASSIVE;
 		}
 	} else if (msg->type == CPC_MSG_TYPE_OVERRUN) {
-		cf->can_id |= CAN_ERR_CRTL;
-		cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW;
+		if (skb) {
+			cf->can_id |= CAN_ERR_CRTL;
+			cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW;
+		}
 
 		stats->rx_over_errors++;
 		stats->rx_errors++;
 	}
 
-	netif_rx(skb);
+	if (skb)
+		netif_rx(skb);
 }
 
 /*

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #16 lore raw

PATCH 16/356 can: f81604: f81604_handle_can_bus_errors(): fix {rx,tx}_errors statistics

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dario Binacchi <[email protected]>

[ Upstream commit d7b916540c2ba3d2a88c27b2a6287b39d8eac052 ]

The f81604_handle_can_bus_errors() function only incremented the receive error counter and never the transmit error counter, even if the ECC_DIR flag reported that an error had occurred during transmission.

Increment the receive/transmit error counter based on the value of the ECC_DIR flag.

Fixes 88da17436973 ("can: usb: f81604: add Fintek F81604 support")

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/can/usb/f81604.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

drivers/net/can/usb/f81604.c +7 -3

index ec8cef7fd2d53..fb9fb16507f0b 100644
@@ -526,7 +526,6 @@ static void f81604_handle_can_bus_errors(struct f81604_port_priv *priv,
 		netdev_dbg(netdev, "bus error interrupt\n");
 
 		priv->can.can_stats.bus_error++;
-		stats->rx_errors++;
 
 		if (skb) {
 			cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;

@@ -548,10 +547,15 @@ static void f81604_handle_can_bus_errors(struct f81604_port_priv *priv,
 
 			/* set error location */
 			cf->data[3] = data->ecc & F81604_SJA1000_ECC_SEG;
+		}
 
-			/* Error occurred during transmission? */
-			if ((data->ecc & F81604_SJA1000_ECC_DIR) == 0)
+		/* Error occurred during transmission? */
+		if ((data->ecc & F81604_SJA1000_ECC_DIR) == 0) {
+			stats->tx_errors++;
+			if (skb)
 				cf->data[2] |= CAN_ERR_PROT_TX;
+		} else {
+			stats->rx_errors++;
 		}
 
 		set_bit(F81604_CLEAR_ECC, &priv->clear_flags);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #17 lore raw

PATCH 17/356 ipvs: fix UB due to uninitialized stack access in ip_vs_protocol_init()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jinghao Jia <[email protected]>

[ Upstream commit 146b6f1112eb30a19776d6c323c994e9d67790db ]

Under certain kernel configurations when building with Clang/LLVM, the compiler does not generate a return or jump as the terminator instruction for ip_vs_protocol_init(), triggering the following objtool warning during build time:

vmlinux.o: warning: objtool: ip_vs_protocol_init() falls through to next function __initstub__kmod_ip_vs_rr__935_123_ip_vs_rr_init6()

At runtime, this either causes an oops when trying to load the ipvs module or a boot-time panic if ipvs is built-in. This same issue has been reported by the Intel kernel test robot previously.

Digging deeper into both LLVM and the kernel code reveals this to be a undefined behavior problem. ip_vs_protocol_init() uses a on-stack buffer of 64 chars to store the registered protocol names and leaves it uninitialized after definition. The function calls strnlen() when concatenating protocol names into the buffer. With CONFIG_FORTIFY_SOURCE strnlen() performs an extra step to check whether the last byte of the input char buffer is a null character (commit 3009f891bb9f ("fortify:
Allow strlen() and strnlen() to pass compile-time known lengths")). This, together with possibly other configurations, cause the following IR to be generated:

  define hidden i32 @ip_vs_protocol_init() local_unnamed_addr #5 section ".init.text" align 16 !kcfi_type !29 {
    %1 = alloca [64 x i8], align 16
    ...

14: ; preds = %11

    %15 = getelementptr inbounds i8, ptr %1, i64 63
    %16 = load i8, ptr %15, align 1
    %17 = tail call i1 @llvm.is.constant.i8(i8 %16)
    %18 = icmp eq i8 %16, 0
    %19 = select i1 %17, i1 %18, i1 false
    br i1 %19, label %20, label %23

20: ; preds = %14
%21 = call i64 @strlen(ptr noundef nonnull dereferenceable(1) %1) #23
...

  23:                                               ; preds = %14, %11, %20
    %24 = call i64 @strnlen(ptr noundef nonnull dereferenceable(1) %1, i64 noundef 64) #24
    ...
  }

The above code calculates the address of the last char in the buffer (value %15) and then loads from it (value %16). Because the buffer is never initialized, the LLVM GVN pass marks value %16 as undefined:

%13 = getelementptr inbounds i8, ptr %1, i64 63
br i1 undef, label %14, label %17

This gives later passes (SCCP, in particular) more DCE opportunities by propagating the undef value further, and eventually removes everything after the load on the uninitialized stack location:

  define hidden i32 @ip_vs_protocol_init() local_unnamed_addr #0 section ".init.text" align 16 !kcfi_type !11 {
    %1 = alloca [64 x i8], align 16
    ...

  12:                                               ; preds = %11
    %13 = getelementptr inbounds i8, ptr %1, i64 63
    unreachable
  }

In this way, the generated native code will just fall through to the next function, as LLVM does not generate any code for the unreachable IR instruction and leaves the function without a terminator.

Zero the on-stack buffer to avoid this possible UB.

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")

Reported-by kernel test robot <[email protected]>

Closes https://lore.kernel.org/oe-kbuild-all/[email protected]/ open here

Co-developed-by Ruowen Qin <[email protected]>

Signed-off-by Ruowen Qin <[email protected]>

Signed-off-by Jinghao Jia <[email protected]>

Acked-by Julian Anastasov <[email protected]>

Signed-off-by Pablo Neira Ayuso <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/netfilter/ipvs/ip_vs_proto.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

net/netfilter/ipvs/ip_vs_proto.c +1 -3

index f100da4ba3bc3..a9fd1d3fc2cbf 100644
@@ -340,7 +340,7 @@ void __net_exit ip_vs_protocol_net_cleanup(struct netns_ipvs *ipvs)
 
 int __init ip_vs_protocol_init(void)
 {
-	char protocols[64];
+	char protocols[64] = { 0 };
 #define REGISTER_PROTOCOL(p)			\
 	do {					\
 		register_ip_vs_protocol(p);	\

@@ -348,8 +348,6 @@ int __init ip_vs_protocol_init(void)
 		strcat(protocols, (p)->name);	\
 	} while (0)
 
-	protocols[0] = '\0';
-	protocols[2] = '\0';
 #ifdef CONFIG_IP_VS_PROTO_TCP
 	REGISTER_PROTOCOL(&ip_vs_protocol_tcp);
 #endif

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #18 lore raw

PATCH 18/356 netfilter: x_tables: fix LED ID check in led_tg_check()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dmitry Antipov <[email protected]>

[ Upstream commit 04317f4eb2aad312ad85c1a17ad81fe75f1f9bc7 ]

Syzbot has reported the following BUG detected by KASAN:

BUG: KASAN: slab-out-of-bounds in strlen+0x58/0x70
Read of size 1 at addr ffff8881022da0c8 by task repro/5879
...
Call Trace:
 <TASK>
 dump_stack_lvl+0x241/0x360
 ? __pfx_dump_stack_lvl+0x10/0x10
 ? __pfx__printk+0x10/0x10
 ? _printk+0xd5/0x120
 ? __virt_addr_valid+0x183/0x530
 ? __virt_addr_valid+0x183/0x530
 print_report+0x169/0x550
 ? __virt_addr_valid+0x183/0x530
 ? __virt_addr_valid+0x183/0x530
 ? __virt_addr_valid+0x45f/0x530
 ? __phys_addr+0xba/0x170
 ? strlen+0x58/0x70
 kasan_report+0x143/0x180
 ? strlen+0x58/0x70
 strlen+0x58/0x70
 kstrdup+0x20/0x80
 led_tg_check+0x18b/0x3c0
 xt_check_target+0x3bb/0xa40
 ? __pfx_xt_check_target+0x10/0x10
 ? stack_depot_save_flags+0x6e4/0x830
 ? nft_target_init+0x174/0xc30
 nft_target_init+0x82d/0xc30
 ? __pfx_nft_target_init+0x10/0x10
 ? nf_tables_newrule+0x1609/0x2980
 ? nf_tables_newrule+0x1609/0x2980
 ? rcu_is_watching+0x15/0xb0
 ? nf_tables_newrule+0x1609/0x2980
 ? nf_tables_newrule+0x1609/0x2980
 ? __kmalloc_noprof+0x21a/0x400
 nf_tables_newrule+0x1860/0x2980
 ? __pfx_nf_tables_newrule+0x10/0x10
 ? __nla_parse+0x40/0x60
 nfnetlink_rcv+0x14e5/0x2ab0
 ? __pfx_validate_chain+0x10/0x10
 ? __pfx_nfnetlink_rcv+0x10/0x10
 ? __lock_acquire+0x1384/0x2050
 ? netlink_deliver_tap+0x2e/0x1b0
 ? __pfx_lock_release+0x10/0x10
 ? netlink_deliver_tap+0x2e/0x1b0
 netlink_unicast+0x7f8/0x990
 ? __pfx_netlink_unicast+0x10/0x10
 ? __virt_addr_valid+0x183/0x530
 ? __check_object_size+0x48e/0x900
 netlink_sendmsg+0x8e4/0xcb0
 ? __pfx_netlink_sendmsg+0x10/0x10
 ? aa_sock_msg_perm+0x91/0x160
 ? __pfx_netlink_sendmsg+0x10/0x10
 __sock_sendmsg+0x223/0x270
 ____sys_sendmsg+0x52a/0x7e0
 ? __pfx_____sys_sendmsg+0x10/0x10
 __sys_sendmsg+0x292/0x380
 ? __pfx___sys_sendmsg+0x10/0x10
 ? lockdep_hardirqs_on_prepare+0x43d/0x780
 ? __pfx_lockdep_hardirqs_on_prepare+0x10/0x10
 ? exc_page_fault+0x590/0x8c0
 ? do_syscall_64+0xb6/0x230
 do_syscall_64+0xf3/0x230
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
...
 </TASK>

Since an invalid (without '\0' byte at all) byte sequence may be passed from userspace, add an extra check to ensure that such a sequence is rejected as possible ID and so never passed to 'kstrdup()' and further.

Reported-by [email protected]

Closes https://syzkaller.appspot.com/bug?extid=6c8215822f35fdb35667

Fixes 268cb38e1802 ("netfilter: x_tables: add LED trigger target")

Signed-off-by Dmitry Antipov <[email protected]>

Signed-off-by Pablo Neira Ayuso <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/netfilter/xt_LED.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

net/netfilter/xt_LED.c +3 -1

index f7b0286d106ac..8a80fd76fe45b 100644
@@ -96,7 +96,9 @@ static int led_tg_check(const struct xt_tgchk_param *par)
 	struct xt_led_info_internal *ledinternal;
 	int err;
 
-	if (ledinfo->id[0] == '\0')
+	/* Bail out if empty string or not a string at all. */
+	if (ledinfo->id[0] == '\0' ||
+	    !memchr(ledinfo->id, '\0', sizeof(ledinfo->id)))
 		return -EINVAL;
 
 	mutex_lock(&xt_led_mutex);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:30 UTC #19 lore raw

PATCH 19/356 netfilter: nft_socket: remove WARN_ON_ONCE on maximum cgroup level

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit b7529880cb961d515642ce63f9d7570869bbbdc3 ]

cgroup maximum depth is INT_MAX by default, there is a cgroup toggle to restrict this maximum depth to a more reasonable value not to harm performance. Remove unnecessary WARN_ON_ONCE which is reachable from userspace.

Fixes 7f3287db6543 ("netfilter: nft_socket: make cgroupsv2 matching work with namespaces")

Reported-by [email protected]

Signed-off-by Pablo Neira Ayuso <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/netfilter/nft_socket.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/netfilter/nft_socket.c +1 -1

index 0a8883a93e836..187b667bad6c3 100644
@@ -68,7 +68,7 @@ static noinline int nft_socket_cgroup_subtree_level(void)
 
 	cgroup_put(cgrp);
 
-	if (WARN_ON_ONCE(level > 255))
+	if (level > 255)
 		return -ERANGE;
 
 	if (WARN_ON_ONCE(level < 0))

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #20 lore raw

PATCH 20/356 selftests: hid: fix typo and exit code

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maximilian Heyne <[email protected]>

[ Upstream commit e8f34747bddedaf3895e5d5066e0f71713fff811 ]

The correct exit code to mark a test as skipped is 4.

Fixes ffb85d5c9e80 ("selftests: hid: import hid-tools hid-core tests")

Signed-off-by Maximilian Heyne <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Benjamin Tissoires <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 .../testing/selftests/hid/run-hid-tools-tests.sh | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

tools/testing/selftests/hid/run-hid-tools-tests.sh +9 -7

index bdae8464da865..af1682a53c27e 100755
@@ -2,24 +2,26 @@
 # SPDX-License-Identifier: GPL-2.0
 # Runs tests for the HID subsystem
 
+KSELFTEST_SKIP_TEST=4
+
 if ! command -v python3 > /dev/null 2>&1; then
 	echo "hid-tools: [SKIP] python3 not installed"
-	exit 77
+	exit $KSELFTEST_SKIP_TEST
 fi
 
 if ! python3 -c "import pytest" > /dev/null 2>&1; then
-	echo "hid: [SKIP/ pytest module not installed"
-	exit 77
+	echo "hid: [SKIP] pytest module not installed"
+	exit $KSELFTEST_SKIP_TEST
 fi
 
 if ! python3 -c "import pytest_tap" > /dev/null 2>&1; then
-	echo "hid: [SKIP/ pytest_tap module not installed"
-	exit 77
+	echo "hid: [SKIP] pytest_tap module not installed"
+	exit $KSELFTEST_SKIP_TEST
 fi
 
 if ! python3 -c "import hidtools" > /dev/null 2>&1; then
-	echo "hid: [SKIP/ hid-tools module not installed"
-	exit 77
+	echo "hid: [SKIP] hid-tools module not installed"
+	exit $KSELFTEST_SKIP_TEST
 fi
 
 TARGET=${TARGET:=.}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #21 lore raw

PATCH 21/356 net: enetc: Do not configure preemptible TCs if SIs do not support

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wei Fang <[email protected]>

[ Upstream commit b2420b8c81ec674552d00c55d46245e5c184b260 ]

Both ENETC PF and VF drivers share enetc_setup_tc_mqprio() to configure MQPRIO. And enetc_setup_tc_mqprio() calls enetc_change_preemptible_tcs() to configure preemptible TCs. However, only PF is able to configure preemptible TCs. Because only PF has related registers, while VF does not have these registers. So for VF, its hw->port pointer is NULL. Therefore, VF will access an invalid pointer when accessing a non-existent register, which will cause a crash issue. The simplified log is as follows.

root@ls1028ardb:~# tc qdisc add dev eno0vf0 parent root handle 100: \
mqprio num_tc 4 map 0 0 1 1 2 2 3 3 queues 1@0 1@1 1@2 1@3 hw 1
[  187.290775] Unable to handle kernel paging request at virtual address 0000000000001f00
[  187.424831] pc : enetc_mm_commit_preemptible_tcs+0x1c4/0x400
[  187.430518] lr : enetc_mm_commit_preemptible_tcs+0x30c/0x400
[  187.511140] Call trace:
[  187.513588]  enetc_mm_commit_preemptible_tcs+0x1c4/0x400
[  187.518918]  enetc_setup_tc_mqprio+0x180/0x214
[  187.523374]  enetc_vf_setup_tc+0x1c/0x30
[  187.527306]  mqprio_enable_offload+0x144/0x178
[  187.531766]  mqprio_init+0x3ec/0x668
[  187.535351]  qdisc_create+0x15c/0x488
[  187.539023]  tc_modify_qdisc+0x398/0x73c
[  187.542958]  rtnetlink_rcv_msg+0x128/0x378
[  187.547064]  netlink_rcv_skb+0x60/0x130
[  187.550910]  rtnetlink_rcv+0x18/0x24
[  187.554492]  netlink_unicast+0x300/0x36c
[  187.558425]  netlink_sendmsg+0x1a8/0x420
[  187.606759] ---[ end trace 0000000000000000 ]---

In addition, some PFs also do not support configuring preemptible TCs, such as eno1 and eno3 on LS1028A. It won't crash like it does for VFs, but we should prevent these PFs from accessing these unimplemented registers.

Fixes 827145392a4a ("net: enetc: only commit preemptible TCs to hardware when MM TX is active")

Signed-off-by Wei Fang <[email protected]>

Reviewed-by Frank Li <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/freescale/enetc/enetc.c | 3 +++
 1 file changed, 3 insertions(+)

drivers/net/ethernet/freescale/enetc/enetc.c +3 -0

index c17b9e3385168..87b27bd7a13bb 100644
@@ -28,6 +28,9 @@ EXPORT_SYMBOL_GPL(enetc_port_mac_wr);
 static void enetc_change_preemptible_tcs(struct enetc_ndev_priv *priv,
 					 u8 preemptible_tcs)
 {
+	if (!(priv->si->hw_features & ENETC_SI_F_QBU))
+		return;
+
 	priv->preemptible_tcs = preemptible_tcs;
 	enetc_mm_commit_preemptible_tcs(priv);
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #22 lore raw

PATCH 22/356 ptp: Add error handling for adjfine callback in ptp_clock_adjtime

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ajay Kaher <[email protected]>

[ Upstream commit 98337d7c87577ded71114f6976edb70a163e27bc ]

ptp_clock_adjtime sets ptp->dialed_frequency even when adjfine callback returns an error. This causes subsequent reads to return an incorrect value.

Fix this by adding error check before ptp->dialed_frequency is set.

Fixes 39a8cbd9ca05 ("ptp: remember the adjusted frequency")

Signed-off-by Ajay Kaher <[email protected]>

Acked-by Richard Cochran <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/ptp/ptp_clock.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/ptp/ptp_clock.c +2 -1

index 9a50bfb56453c..b586da2e30023 100644
@@ -133,7 +133,8 @@ static int ptp_clock_adjtime(struct posix_clock *pc, struct __kernel_timex *tx)
 		if (ppb > ops->max_adj || ppb < -ops->max_adj)
 			return -ERANGE;
 		err = ops->adjfine(ops, tx->freq);
-		ptp->dialed_frequency = tx->freq;
+		if (!err)
+			ptp->dialed_frequency = tx->freq;
 	} else if (tx->modes & ADJ_OFFSET) {
 		if (ops->adjphase) {
 			s32 max_phase_adj = ops->getmaxphase(ops);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #23 lore raw

PATCH 23/356 net/sched: tbf: correct backlog statistic for GSO packets

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Martin Ottens <[email protected]>

[ Upstream commit 1596a135e3180c92e42dd1fbcad321f4fb3e3b17 ]

When the length of a GSO packet in the tbf qdisc is larger than the burst size configured the packet will be segmented by the tbf_segment function. Whenever this function is used to enqueue SKBs, the backlog statistic of the tbf is not increased correctly. This can lead to underflows of the 'backlog' byte-statistic value when these packets are dequeued from tbf.

Reproduce the bug:
Ensure that the sender machine has GSO enabled. Configured the tbf on the outgoing interface of the machine as follows (burstsize = 1 MTU):

$ tc qdisc add dev <oif> root handle 1: tbf rate 50Mbit burst 1514 latency 50ms

Send bulk TCP traffic out via this interface, e.g., by running an iPerf3 client on this machine. Check the qdisc statistics:

$ tc -s qdisc show dev <oif>

The 'backlog' byte-statistic has incorrect values while traffic is transferred, e.g., high values due to u32 underflows. When the transfer is stopped, the value is != 0, which should never happen.

This patch fixes this bug by updating the statistics correctly, even if single SKBs of a GSO SKB cannot be enqueued.

Fixes e43ac79a4bc6 ("sch_tbf: segment too big GSO packets")

Signed-off-by Martin Ottens <[email protected]>

Reviewed-by Eric Dumazet <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 net/sched/sch_tbf.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

net/sched/sch_tbf.c +12 -6

index 17d2d00ddb182..f92174008499b 100644
@@ -208,7 +208,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch,
 	struct tbf_sched_data *q = qdisc_priv(sch);
 	struct sk_buff *segs, *nskb;
 	netdev_features_t features = netif_skb_features(skb);
-	unsigned int len = 0, prev_len = qdisc_pkt_len(skb);
+	unsigned int len = 0, prev_len = qdisc_pkt_len(skb), seg_len;
 	int ret, nb;
 
 	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);

@@ -219,21 +219,27 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch,
 	nb = 0;
 	skb_list_walk_safe(segs, segs, nskb) {
 		skb_mark_not_on_list(segs);
-		qdisc_skb_cb(segs)->pkt_len = segs->len;
-		len += segs->len;
+		seg_len = segs->len;
+		qdisc_skb_cb(segs)->pkt_len = seg_len;
 		ret = qdisc_enqueue(segs, q->qdisc, to_free);
 		if (ret != NET_XMIT_SUCCESS) {
 			if (net_xmit_drop_count(ret))
 				qdisc_qstats_drop(sch);
 		} else {
 			nb++;
+			len += seg_len;
 		}
 	}
 	sch->q.qlen += nb;
-	if (nb > 1)
+	sch->qstats.backlog += len;
+	if (nb > 0) {
 		qdisc_tree_reduce_backlog(sch, 1 - nb, prev_len - len);
-	consume_skb(skb);
-	return nb > 0 ? NET_XMIT_SUCCESS : NET_XMIT_DROP;
+		consume_skb(skb);
+		return NET_XMIT_SUCCESS;
+	}
+
+	kfree_skb(skb);
+	return NET_XMIT_DROP;
 }
 
 static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch,

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #24 lore raw

PATCH 24/356 net: hsr: avoid potential out-of-bound access in fill_frame_info()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit b9653d19e556c6afd035602927a93d100a0d7644 ]

syzbot is able to feed a packet with 14 bytes, pretending it is a vlan one.

Since fill_frame_info() is relying on skb->mac_len already, extend the check to cover this case.

BUG: KMSAN: uninit-value in fill_frame_info net/hsr/hsr_forward.c:709 [inline]
 BUG: KMSAN: uninit-value in hsr_forward_skb+0x9ee/0x3b10 net/hsr/hsr_forward.c:724
  fill_frame_info net/hsr/hsr_forward.c:709 [inline]
  hsr_forward_skb+0x9ee/0x3b10 net/hsr/hsr_forward.c:724
  hsr_dev_xmit+0x2f0/0x350 net/hsr/hsr_device.c:235
  __netdev_start_xmit include/linux/netdevice.h:5002 [inline]
  netdev_start_xmit include/linux/netdevice.h:5011 [inline]
  xmit_one net/core/dev.c:3590 [inline]
  dev_hard_start_xmit+0x247/0xa20 net/core/dev.c:3606
  __dev_queue_xmit+0x366a/0x57d0 net/core/dev.c:4434
  dev_queue_xmit include/linux/netdevice.h:3168 [inline]
  packet_xmit+0x9c/0x6c0 net/packet/af_packet.c:276
  packet_snd net/packet/af_packet.c:3146 [inline]
  packet_sendmsg+0x91ae/0xa6f0 net/packet/af_packet.c:3178
  sock_sendmsg_nosec net/socket.c:711 [inline]
  __sock_sendmsg+0x30f/0x380 net/socket.c:726
  __sys_sendto+0x594/0x750 net/socket.c:2197
  __do_sys_sendto net/socket.c:2204 [inline]
  __se_sys_sendto net/socket.c:2200 [inline]
  __x64_sys_sendto+0x125/0x1d0 net/socket.c:2200
  x64_sys_call+0x346a/0x3c30 arch/x86/include/generated/asm/syscalls_64.h:45
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xcd/0x1e0 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Uninit was created at:

  slab_post_alloc_hook mm/slub.c:4091 [inline]
  slab_alloc_node mm/slub.c:4134 [inline]
  kmem_cache_alloc_node_noprof+0x6bf/0xb80 mm/slub.c:4186
  kmalloc_reserve+0x13d/0x4a0 net/core/skbuff.c:587
  __alloc_skb+0x363/0x7b0 net/core/skbuff.c:678
  alloc_skb include/linux/skbuff.h:1323 [inline]
  alloc_skb_with_frags+0xc8/0xd00 net/core/skbuff.c:6612
  sock_alloc_send_pskb+0xa81/0xbf0 net/core/sock.c:2881
  packet_alloc_skb net/packet/af_packet.c:2995 [inline]
  packet_snd net/packet/af_packet.c:3089 [inline]
  packet_sendmsg+0x74c6/0xa6f0 net/packet/af_packet.c:3178
  sock_sendmsg_nosec net/socket.c:711 [inline]
  __sock_sendmsg+0x30f/0x380 net/socket.c:726
  __sys_sendto+0x594/0x750 net/socket.c:2197
  __do_sys_sendto net/socket.c:2204 [inline]
  __se_sys_sendto net/socket.c:2200 [inline]
  __x64_sys_sendto+0x125/0x1d0 net/socket.c:2200
  x64_sys_call+0x346a/0x3c30 arch/x86/include/generated/asm/syscalls_64.h:45
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xcd/0x1e0 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Fixes 48b491a5cc74 ("net: hsr: fix mac_len checks")

Reported-by [email protected]

Closes https://lore.kernel.org/netdev/[email protected]/T/#u open here

Cc WingMan Kwok <[email protected]>

Cc Murali Karicheri <[email protected]>

Cc MD Danish Anwar <[email protected]>

Cc Jiri Pirko <[email protected]>

Cc George McCollister <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 net/hsr/hsr_forward.c | 2 ++
 1 file changed, 2 insertions(+)

net/hsr/hsr_forward.c +2 -0

index 0323ab5023c69..2790f3964d6bd 100644
@@ -588,6 +588,8 @@ static int fill_frame_info(struct hsr_frame_info *frame,
 		frame->is_vlan = true;
 
 	if (frame->is_vlan) {
+		if (skb->mac_len < offsetofend(struct hsr_vlan_ethhdr, vlanhdr))
+			return -EINVAL;
 		vlan_hdr = (struct hsr_vlan_ethhdr *)ethhdr;
 		proto = vlan_hdr->vlanhdr.h_vlan_encapsulated_proto;
 		/* FIXME: */

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #25 lore raw

PATCH 25/356 can: j1939: j1939_session_new(): fix skb reference counting

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dmitry Antipov <[email protected]>

[ Upstream commit a8c695005bfe6569acd73d777ca298ddddd66105 ]

Since j1939_session_skb_queue() does an extra skb_get() for each new skb, do the same for the initial one in j1939_session_new() to avoid refcount underflow.

Reported-by [email protected]

Closes https://syzkaller.appspot.com/bug?extid=d4e8dc385d9258220c31

Fixes 9d71dd0c7009 ("can: add support of SAE J1939 protocol")

Signed-off-by Dmitry Antipov <[email protected]>

Tested-by Oleksij Rempel <[email protected]>

Acked-by Oleksij Rempel <[email protected]>

Link https://patch.msgid.link/[email protected] open here

[mkl: clean up commit message]

Signed-off-by Marc Kleine-Budde <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/can/j1939/transport.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/can/j1939/transport.c +1 -1

index 319f47df33300..95f7a7e65a73f 100644
@@ -1505,7 +1505,7 @@ static struct j1939_session *j1939_session_new(struct j1939_priv *priv,
 	session->state = J1939_SESSION_NEW;
 
 	skb_queue_head_init(&session->skb_queue);
-	skb_queue_tail(&session->skb_queue, skb);
+	skb_queue_tail(&session->skb_queue, skb_get(skb));
 
 	skcb = j1939_skb_to_cb(skb);
 	memcpy(&session->skcb, skcb, sizeof(session->skcb));

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #26 lore raw

PATCH 26/356 platform/x86: asus-wmi: add support for vivobook fan profiles

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mohamed Ghanmi <[email protected]>

[ Upstream commit bcbfcebda2cbc6a10a347d726e4a4f69e43a864e ]

Add support for vivobook fan profiles wmi call on the ASUS VIVOBOOK to adjust power limits.

These fan profiles have a different device id than the ROG series and different order. This reorders the existing modes.

As part of keeping the patch clean the throttle_thermal_policy_available boolean stored in the driver struct is removed and throttle_thermal_policy_dev is used in place (as on init it is zeroed).

Co-developed-by Luke D. Jones <[email protected]>

Signed-off-by Luke D. Jones <[email protected]>

Signed-off-by Mohamed Ghanmi <[email protected]>

Reviewed-by Luke D. Jones <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Reviewed-by Hans de Goede <[email protected]>

Signed-off-by Hans de Goede <[email protected]>

Stable-dep-of: 25fb5f47f34d ("platform/x86: asus-wmi: Ignore return value when writing thermal policy")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/platform/x86/asus-wmi.c            | 120 ++++++++++++---------
 include/linux/platform_data/x86/asus-wmi.h |   1 +
 2 files changed, 71 insertions(+), 50 deletions(-)

drivers/platform/x86/asus-wmi.c +70 ~~-50~~

index 9c6321c2fc3c5..d0ba8bd83fc3d 100644
@@ -95,6 +95,12 @@ module_param(fnlock_default, bool, 0444);
 #define ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST	1
 #define ASUS_THROTTLE_THERMAL_POLICY_SILENT	2
 
+#define ASUS_THROTTLE_THERMAL_POLICY_DEFAULT_VIVO	0
+#define ASUS_THROTTLE_THERMAL_POLICY_SILENT_VIVO	1
+#define ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST_VIVO	2
+
+#define PLATFORM_PROFILE_MAX 2
+
 #define USB_INTEL_XUSB2PR		0xD0
 #define PCI_DEVICE_ID_INTEL_LYNXPOINT_LP_XHCI	0x9c31

@@ -269,8 +275,8 @@ struct asus_wmi {
 	bool kbd_rgb_mode_available;
 	bool kbd_rgb_state_available;
 
-	bool throttle_thermal_policy_available;
 	u8 throttle_thermal_policy_mode;
+	u32 throttle_thermal_policy_dev;
 
 	bool cpu_fan_curve_available;
 	bool gpu_fan_curve_available;

@@ -2912,7 +2918,7 @@ static int fan_curve_get_factory_default(struct asus_wmi *asus, u32 fan_dev)
 	int err, fan_idx;
 	u8 mode = 0;
 
-	if (asus->throttle_thermal_policy_available)
+	if (asus->throttle_thermal_policy_dev)
 		mode = asus->throttle_thermal_policy_mode;
 	/* DEVID_<C/G>PU_FAN_CURVE is switched for OVERBOOST vs SILENT */
 	if (mode == 2)

@@ -3119,7 +3125,7 @@ static ssize_t fan_curve_enable_store(struct device *dev,
 		 * For machines with throttle this is the only way to reset fans
 		 * to default mode of operation (does not erase curve data).
 		 */
-		if (asus->throttle_thermal_policy_available) {
+		if (asus->throttle_thermal_policy_dev) {
 			err = throttle_thermal_policy_write(asus);
 			if (err)
 				return err;

@@ -3336,8 +3342,8 @@ static const struct attribute_group asus_fan_curve_attr_group = {
 __ATTRIBUTE_GROUPS(asus_fan_curve_attr);
 
 /*
- * Must be initialised after throttle_thermal_policy_check_present() as
- * we check the status of throttle_thermal_policy_available during init.
+ * Must be initialised after throttle_thermal_policy_dev is set as
+ * we check the status of throttle_thermal_policy_dev during init.
  */
 static int asus_wmi_custom_fan_curve_init(struct asus_wmi *asus)
 {

@@ -3378,38 +3384,13 @@ static int asus_wmi_custom_fan_curve_init(struct asus_wmi *asus)
 }
 
 /* Throttle thermal policy ****************************************************/
-
-static int throttle_thermal_policy_check_present(struct asus_wmi *asus)
-{
-	u32 result;
-	int err;
-
-	asus->throttle_thermal_policy_available = false;
-
-	err = asus_wmi_get_devstate(asus,
-				    ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY,
-				    &result);
-	if (err) {
-		if (err == -ENODEV)
-			return 0;
-		return err;
-	}
-
-	if (result & ASUS_WMI_DSTS_PRESENCE_BIT)
-		asus->throttle_thermal_policy_available = true;
-
-	return 0;
-}
-
 static int throttle_thermal_policy_write(struct asus_wmi *asus)
 {
-	int err;
-	u8 value;
+	u8 value = asus->throttle_thermal_policy_mode;
 	u32 retval;
+	int err;
 
-	value = asus->throttle_thermal_policy_mode;
-
-	err = asus_wmi_set_devstate(ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY,
+	err = asus_wmi_set_devstate(asus->throttle_thermal_policy_dev,
 				    value, &retval);
 
 	sysfs_notify(&asus->platform_device->dev.kobj, NULL,

@@ -3439,7 +3420,7 @@ static int throttle_thermal_policy_write(struct asus_wmi *asus)
 
 static int throttle_thermal_policy_set_default(struct asus_wmi *asus)
 {
-	if (!asus->throttle_thermal_policy_available)
+	if (!asus->throttle_thermal_policy_dev)
 		return 0;
 
 	asus->throttle_thermal_policy_mode = ASUS_THROTTLE_THERMAL_POLICY_DEFAULT;

@@ -3451,7 +3432,7 @@ static int throttle_thermal_policy_switch_next(struct asus_wmi *asus)
 	u8 new_mode = asus->throttle_thermal_policy_mode + 1;
 	int err;
 
-	if (new_mode > ASUS_THROTTLE_THERMAL_POLICY_SILENT)
+	if (new_mode > PLATFORM_PROFILE_MAX)
 		new_mode = ASUS_THROTTLE_THERMAL_POLICY_DEFAULT;
 
 	asus->throttle_thermal_policy_mode = new_mode;

@@ -3490,7 +3471,7 @@ static ssize_t throttle_thermal_policy_store(struct device *dev,
 	if (result < 0)
 		return result;
 
-	if (new_mode > ASUS_THROTTLE_THERMAL_POLICY_SILENT)
+	if (new_mode > PLATFORM_PROFILE_MAX)
 		return -EINVAL;
 
 	asus->throttle_thermal_policy_mode = new_mode;

@@ -3507,10 +3488,52 @@ static ssize_t throttle_thermal_policy_store(struct device *dev,
 	return count;
 }
 
-// Throttle thermal policy: 0 - default, 1 - overboost, 2 - silent
+/*
+ * Throttle thermal policy: 0 - default, 1 - overboost, 2 - silent
+ */
 static DEVICE_ATTR_RW(throttle_thermal_policy);
 
 /* Platform profile ***********************************************************/
+static int asus_wmi_platform_profile_to_vivo(struct asus_wmi *asus, int mode)
+{
+	bool vivo;
+
+	vivo = asus->throttle_thermal_policy_dev == ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY_VIVO;
+
+	if (vivo) {
+		switch (mode) {
+		case ASUS_THROTTLE_THERMAL_POLICY_DEFAULT:
+			return ASUS_THROTTLE_THERMAL_POLICY_DEFAULT_VIVO;
+		case ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST:
+			return ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST_VIVO;
+		case ASUS_THROTTLE_THERMAL_POLICY_SILENT:
+			return ASUS_THROTTLE_THERMAL_POLICY_SILENT_VIVO;
+		}
+	}
+
+	return mode;
+}
+
+static int asus_wmi_platform_profile_mode_from_vivo(struct asus_wmi *asus, int mode)
+{
+	bool vivo;
+
+	vivo = asus->throttle_thermal_policy_dev == ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY_VIVO;
+
+	if (vivo) {
+		switch (mode) {
+		case ASUS_THROTTLE_THERMAL_POLICY_DEFAULT_VIVO:
+			return ASUS_THROTTLE_THERMAL_POLICY_DEFAULT;
+		case ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST_VIVO:
+			return ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST;
+		case ASUS_THROTTLE_THERMAL_POLICY_SILENT_VIVO:
+			return ASUS_THROTTLE_THERMAL_POLICY_SILENT;
+		}
+	}
+
+	return mode;
+}
+
 static int asus_wmi_platform_profile_get(struct platform_profile_handler *pprof,
 					enum platform_profile_option *profile)
 {

@@ -3518,10 +3541,9 @@ static int asus_wmi_platform_profile_get(struct platform_profile_handler *pprof,
 	int tp;
 
 	asus = container_of(pprof, struct asus_wmi, platform_profile_handler);
-
 	tp = asus->throttle_thermal_policy_mode;
 
-	switch (tp) {
+	switch (asus_wmi_platform_profile_mode_from_vivo(asus, tp)) {
 	case ASUS_THROTTLE_THERMAL_POLICY_DEFAULT:
 		*profile = PLATFORM_PROFILE_BALANCED;
 		break;

@@ -3560,7 +3582,7 @@ static int asus_wmi_platform_profile_set(struct platform_profile_handler *pprof,
 		return -EOPNOTSUPP;
 	}
 
-	asus->throttle_thermal_policy_mode = tp;
+	asus->throttle_thermal_policy_mode = asus_wmi_platform_profile_to_vivo(asus, tp);
 	return throttle_thermal_policy_write(asus);
 }

@@ -3573,7 +3595,7 @@ static int platform_profile_setup(struct asus_wmi *asus)
 	 * Not an error if a component platform_profile relies on is unavailable
 	 * so early return, skipping the setup of platform_profile.
 	 */
-	if (!asus->throttle_thermal_policy_available)
+	if (!asus->throttle_thermal_policy_dev)
 		return 0;
 
 	dev_info(dev, "Using throttle_thermal_policy for platform_profile support\n");

@@ -3870,7 +3892,7 @@ static void asus_wmi_handle_event_code(int code, struct asus_wmi *asus)
 	if (code == NOTIFY_KBD_FBM || code == NOTIFY_KBD_TTP) {
 		if (asus->fan_boost_mode_available)
 			fan_boost_mode_switch_next(asus);
-		if (asus->throttle_thermal_policy_available)
+		if (asus->throttle_thermal_policy_dev)
 			throttle_thermal_policy_switch_next(asus);
 		return;

@@ -4075,7 +4097,7 @@ static umode_t asus_sysfs_is_visible(struct kobject *kobj,
 	else if (attr == &dev_attr_fan_boost_mode.attr)
 		ok = asus->fan_boost_mode_available;
 	else if (attr == &dev_attr_throttle_thermal_policy.attr)
-		ok = asus->throttle_thermal_policy_available;
+		ok = asus->throttle_thermal_policy_dev != 0;
 	else if (attr == &dev_attr_ppt_pl2_sppt.attr)
 		ok = asus->ppt_pl2_sppt_available;
 	else if (attr == &dev_attr_ppt_pl1_spl.attr)

@@ -4365,16 +4387,15 @@ static int asus_wmi_add(struct platform_device *pdev)
 	asus->panel_overdrive_available = asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_PANEL_OD);
 	asus->mini_led_mode_available = asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_MINI_LED_MODE);
 
+	if (asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY))
+		asus->throttle_thermal_policy_dev = ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY;
+	else if (asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY_VIVO))
+		asus->throttle_thermal_policy_dev = ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY_VIVO;
+
 	err = fan_boost_mode_check_present(asus);
 	if (err)
 		goto fail_fan_boost_mode;
 
-	err = throttle_thermal_policy_check_present(asus);
-	if (err)
-		goto fail_throttle_thermal_policy;
-	else
-		throttle_thermal_policy_set_default(asus);
-
 	err = platform_profile_setup(asus);
 	if (err)
 		goto fail_platform_profile_setup;

@@ -4461,7 +4482,6 @@ static int asus_wmi_add(struct platform_device *pdev)
 fail_input:
 	asus_wmi_sysfs_exit(asus->platform_device);
 fail_sysfs:
-fail_throttle_thermal_policy:
 fail_custom_fan_curve:
 fail_platform_profile_setup:
 	if (asus->platform_profile_support)

include/linux/platform_data/x86/asus-wmi.h +1 -0

index 16e99a1c37fc4..8e48bdeb55493 100644
@@ -60,6 +60,7 @@
 #define ASUS_WMI_DEVID_LIGHTBAR		0x00050025
 #define ASUS_WMI_DEVID_FAN_BOOST_MODE	0x00110018
 #define ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY 0x00120075
+#define ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY_VIVO 0x00110019
 
 /* Misc */
 #define ASUS_WMI_DEVID_PANEL_OD		0x00050019

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #27 lore raw

PATCH 27/356 platform/x86: asus-wmi: Fix inconsistent use of thermal policies

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Armin Wolf <[email protected]>

[ Upstream commit 895085ec3f2ed7a26389943729e2904df1f88dc0 ]

When changing the thermal policy using the platform profile API, a Vivobook thermal policy is stored in throttle_thermal_policy_mode.

However everywhere else a normal thermal policy is stored inside this variable, potentially confusing the platform profile.

Fix this by always storing normal thermal policy values inside throttle_thermal_policy_mode and only do the conversion when writing the thermal policy to hardware. This also fixes the order in which throttle_thermal_policy_switch_next() steps through the thermal modes on Vivobook machines.

Tested-by Casey G Bowman <[email protected]>

Fixes bcbfcebda2cb ("platform/x86: asus-wmi: add support for vivobook fan profiles")

Signed-off-by Armin Wolf <[email protected]>

Reviewed-by Hans de Goede <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Reviewed-by Ilpo Järvinen <[email protected]>

Signed-off-by Ilpo Järvinen <[email protected]>

Stable-dep-of: 25fb5f47f34d ("platform/x86: asus-wmi: Ignore return value when writing thermal policy")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/platform/x86/asus-wmi.c | 64 +++++++++++----------------------
 1 file changed, 21 insertions(+), 43 deletions(-)

drivers/platform/x86/asus-wmi.c +21 ~~-43~~

index d0ba8bd83fc3d..e36c299dcfb17 100644
@@ -3386,10 +3386,28 @@ static int asus_wmi_custom_fan_curve_init(struct asus_wmi *asus)
 /* Throttle thermal policy ****************************************************/
 static int throttle_thermal_policy_write(struct asus_wmi *asus)
 {
-	u8 value = asus->throttle_thermal_policy_mode;
 	u32 retval;
+	u8 value;
 	int err;
 
+	if (asus->throttle_thermal_policy_dev == ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY_VIVO) {
+		switch (asus->throttle_thermal_policy_mode) {
+		case ASUS_THROTTLE_THERMAL_POLICY_DEFAULT:
+			value = ASUS_THROTTLE_THERMAL_POLICY_DEFAULT_VIVO;
+			break;
+		case ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST:
+			value = ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST_VIVO;
+			break;
+		case ASUS_THROTTLE_THERMAL_POLICY_SILENT:
+			value = ASUS_THROTTLE_THERMAL_POLICY_SILENT_VIVO;
+			break;
+		default:
+			return -EINVAL;
+		}
+	} else {
+		value = asus->throttle_thermal_policy_mode;
+	}
+
 	err = asus_wmi_set_devstate(asus->throttle_thermal_policy_dev,
 				    value, &retval);

@@ -3494,46 +3512,6 @@ static ssize_t throttle_thermal_policy_store(struct device *dev,
 static DEVICE_ATTR_RW(throttle_thermal_policy);
 
 /* Platform profile ***********************************************************/
-static int asus_wmi_platform_profile_to_vivo(struct asus_wmi *asus, int mode)
-{
-	bool vivo;
-
-	vivo = asus->throttle_thermal_policy_dev == ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY_VIVO;
-
-	if (vivo) {
-		switch (mode) {
-		case ASUS_THROTTLE_THERMAL_POLICY_DEFAULT:
-			return ASUS_THROTTLE_THERMAL_POLICY_DEFAULT_VIVO;
-		case ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST:
-			return ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST_VIVO;
-		case ASUS_THROTTLE_THERMAL_POLICY_SILENT:
-			return ASUS_THROTTLE_THERMAL_POLICY_SILENT_VIVO;
-		}
-	}
-
-	return mode;
-}
-
-static int asus_wmi_platform_profile_mode_from_vivo(struct asus_wmi *asus, int mode)
-{
-	bool vivo;
-
-	vivo = asus->throttle_thermal_policy_dev == ASUS_WMI_DEVID_THROTTLE_THERMAL_POLICY_VIVO;
-
-	if (vivo) {
-		switch (mode) {
-		case ASUS_THROTTLE_THERMAL_POLICY_DEFAULT_VIVO:
-			return ASUS_THROTTLE_THERMAL_POLICY_DEFAULT;
-		case ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST_VIVO:
-			return ASUS_THROTTLE_THERMAL_POLICY_OVERBOOST;
-		case ASUS_THROTTLE_THERMAL_POLICY_SILENT_VIVO:
-			return ASUS_THROTTLE_THERMAL_POLICY_SILENT;
-		}
-	}
-
-	return mode;
-}
-
 static int asus_wmi_platform_profile_get(struct platform_profile_handler *pprof,
 					enum platform_profile_option *profile)
 {

@@ -3543,7 +3521,7 @@ static int asus_wmi_platform_profile_get(struct platform_profile_handler *pprof,
 	asus = container_of(pprof, struct asus_wmi, platform_profile_handler);
 	tp = asus->throttle_thermal_policy_mode;
 
-	switch (asus_wmi_platform_profile_mode_from_vivo(asus, tp)) {
+	switch (tp) {
 	case ASUS_THROTTLE_THERMAL_POLICY_DEFAULT:
 		*profile = PLATFORM_PROFILE_BALANCED;
 		break;

@@ -3582,7 +3560,7 @@ static int asus_wmi_platform_profile_set(struct platform_profile_handler *pprof,
 		return -EOPNOTSUPP;
 	}
 
-	asus->throttle_thermal_policy_mode = asus_wmi_platform_profile_to_vivo(asus, tp);
+	asus->throttle_thermal_policy_mode = tp;
 	return throttle_thermal_policy_write(asus);
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #28 lore raw

PATCH 28/356 platform/x86: asus-wmi: Ignore return value when writing thermal policy

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Armin Wolf <[email protected]>

[ Upstream commit 25fb5f47f34d90aceda2c47a4230315536e97fa8 ]

On some machines like the ASUS Vivobook S14 writing the thermal policy returns the currently writen thermal policy instead of an error code.

Ignore the return code to avoid falsely returning an error when the thermal policy was written successfully.

Reported-by [email protected]

Closes https://bugzilla.kernel.org/show_bug.cgi?id=219517

Fixes 2daa86e78c49 ("platform/x86: asus_wmi: Support throttle thermal policy")

Signed-off-by Armin Wolf <[email protected]>

Reviewed-by Hans de Goede <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Reviewed-by Ilpo Järvinen <[email protected]>

Signed-off-by Ilpo Järvinen <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/platform/x86/asus-wmi.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

drivers/platform/x86/asus-wmi.c +2 -9

index e36c299dcfb17..1bf6178a3a105 100644
@@ -3386,7 +3386,6 @@ static int asus_wmi_custom_fan_curve_init(struct asus_wmi *asus)
 /* Throttle thermal policy ****************************************************/
 static int throttle_thermal_policy_write(struct asus_wmi *asus)
 {
-	u32 retval;
 	u8 value;
 	int err;

@@ -3408,8 +3407,8 @@ static int throttle_thermal_policy_write(struct asus_wmi *asus)
 		value = asus->throttle_thermal_policy_mode;
 	}
 
-	err = asus_wmi_set_devstate(asus->throttle_thermal_policy_dev,
-				    value, &retval);
+	/* Some machines do not return an error code as a result, so we ignore it */
+	err = asus_wmi_set_devstate(asus->throttle_thermal_policy_dev, value, NULL);
 
 	sysfs_notify(&asus->platform_device->dev.kobj, NULL,
 			"throttle_thermal_policy");

@@ -3419,12 +3418,6 @@ static int throttle_thermal_policy_write(struct asus_wmi *asus)
 		return err;
 	}
 
-	if (retval != 1) {
-		pr_warn("Failed to set throttle thermal policy (retval): 0x%x\n",
-			retval);
-		return -EIO;
-	}
-
 	/* Must set to disabled if mode is toggled */
 	if (asus->cpu_fan_curve_available)
 		asus->custom_fan_curves[FAN_CURVE_DEV_CPU].enabled = false;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #29 lore raw

PATCH 29/356 net-timestamp: make sk_tskey more predictable in error path

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vadim Fedorenko <[email protected]>

[ Upstream commit 488b6d91b07112eaaaa4454332c1480894d4e06e ]

When SOF_TIMESTAMPING_OPT_ID is used to ambiguate timestamped datagrams, the sk_tskey can become unpredictable in case of any error happened during sendmsg(). Move increment later in the code and make decrement of sk_tskey in error path. This solution is still racy in case of multiple threads doing snedmsg() over the very same socket in parallel, but still makes error path much more predictable.

Fixes 09c2d251b707 ("net-timestamp: add key to disambiguate concurrent datagrams")

Reported-by Andy Lutomirski <[email protected]>

Signed-off-by Vadim Fedorenko <[email protected]>

Reviewed-by Willem de Bruijn <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Paolo Abeni <[email protected]>

Stable-dep-of: 3301ab7d5aeb ("net/ipv6: release expired exception dst cached in socket")

Signed-off-by Sasha Levin <[email protected]>

---
 net/ipv4/ip_output.c  | 13 ++++++++-----
 net/ipv6/ip6_output.c | 13 ++++++++-----
 2 files changed, 16 insertions(+), 10 deletions(-)

net/ipv4/ip_output.c +8 -5

index 2458461e24874..765bd3f2a8408 100644
@@ -972,8 +972,8 @@ static int __ip_append_data(struct sock *sk,
 	unsigned int maxfraglen, fragheaderlen, maxnonfragsize;
 	int csummode = CHECKSUM_NONE;
 	struct rtable *rt = (struct rtable *)cork->dst;
+	bool paged, hold_tskey, extra_uref = false;
 	unsigned int wmem_alloc_delta = 0;
-	bool paged, extra_uref = false;
 	u32 tskey = 0;
 
 	skb = skb_peek_tail(queue);

@@ -982,10 +982,6 @@ static int __ip_append_data(struct sock *sk,
 	mtu = cork->gso_size ? IP_MAX_MTU : cork->fragsize;
 	paged = !!cork->gso_size;
 
-	if (cork->tx_flags & SKBTX_ANY_TSTAMP &&
-	    READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_ID)
-		tskey = atomic_inc_return(&sk->sk_tskey) - 1;
-
 	hh_len = LL_RESERVED_SPACE(rt->dst.dev);
 
 	fragheaderlen = sizeof(struct iphdr) + (opt ? opt->optlen : 0);

@@ -1052,6 +1048,11 @@ static int __ip_append_data(struct sock *sk,
 
 	cork->length += length;
 
+	hold_tskey = cork->tx_flags & SKBTX_ANY_TSTAMP &&
+		     READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_ID;
+	if (hold_tskey)
+		tskey = atomic_inc_return(&sk->sk_tskey) - 1;
+
 	/* So, what's going on in the loop below?
 	 *
 	 * We use calculated fragment length to generate chained skb,

@@ -1274,6 +1275,8 @@ static int __ip_append_data(struct sock *sk,
 	cork->length -= length;
 	IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTDISCARDS);
 	refcount_add(wmem_alloc_delta, &sk->sk_wmem_alloc);
+	if (hold_tskey)
+		atomic_dec(&sk->sk_tskey);
 	return err;
 }

net/ipv6/ip6_output.c +8 -5

index 5d8d86c159dc3..65e2f19814358 100644
@@ -1501,11 +1501,11 @@ static int __ip6_append_data(struct sock *sk,
 	bool zc = false;
 	u32 tskey = 0;
 	struct rt6_info *rt = (struct rt6_info *)cork->dst;
+	bool paged, hold_tskey, extra_uref = false;
 	struct ipv6_txoptions *opt = v6_cork->opt;
 	int csummode = CHECKSUM_NONE;
 	unsigned int maxnonfragsize, headersize;
 	unsigned int wmem_alloc_delta = 0;
-	bool paged, extra_uref = false;
 
 	skb = skb_peek_tail(queue);
 	if (!skb) {

@@ -1517,10 +1517,6 @@ static int __ip6_append_data(struct sock *sk,
 	mtu = cork->gso_size ? IP6_MAX_MTU : cork->fragsize;
 	orig_mtu = mtu;
 
-	if (cork->tx_flags & SKBTX_ANY_TSTAMP &&
-	    READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_ID)
-		tskey = atomic_inc_return(&sk->sk_tskey) - 1;
-
 	hh_len = LL_RESERVED_SPACE(rt->dst.dev);
 
 	fragheaderlen = sizeof(struct ipv6hdr) + rt->rt6i_nfheader_len +

@@ -1617,6 +1613,11 @@ static int __ip6_append_data(struct sock *sk,
 			flags &= ~MSG_SPLICE_PAGES;
 	}
 
+	hold_tskey = cork->tx_flags & SKBTX_ANY_TSTAMP &&
+		     READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_ID;
+	if (hold_tskey)
+		tskey = atomic_inc_return(&sk->sk_tskey) - 1;
+
 	/*
 	 * Let's try using as much space as possible.
 	 * Use MTU if total length of the message fits into the MTU.

@@ -1873,6 +1874,8 @@ static int __ip6_append_data(struct sock *sk,
 	cork->length -= length;
 	IP6_INC_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS);
 	refcount_add(wmem_alloc_delta, &sk->sk_wmem_alloc);
+	if (hold_tskey)
+		atomic_dec(&sk->sk_tskey);
 	return err;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #30 lore raw

PATCH 30/356 ipv6: introduce dst_rt6_info() helper

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit e8dfd42c17faf183415323db1ef0c977be0d6489 ]

Instead of (struct rt6_info *)dst casts, we can use :

#define dst_rt6_info(_ptr) \
container_of_const(_ptr, struct rt6_info, dst)

Some places needed missing const qualifiers :

ip6_confirm_neigh(), ipv6_anycast_destination(),
ipv6_unicast_destination(), has_gateway()

v2: added missing parts (David Ahern)

Reviewed-by David Ahern <[email protected]>

Stable-dep-of: 3301ab7d5aeb ("net/ipv6: release expired exception dst cached in socket")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/infiniband/core/addr.c                |  6 ++--
 .../ethernet/mellanox/mlxsw/spectrum_span.c   |  2 +-
 drivers/net/vrf.c                             |  2 +-
 drivers/net/vxlan/vxlan_core.c                |  2 +-
 drivers/s390/net/qeth_core.h                  |  4 +--
 include/net/ip6_fib.h                         |  6 ++--
 include/net/ip6_route.h                       | 11 ++++----
 net/bluetooth/6lowpan.c                       |  2 +-
 net/core/dst_cache.c                          |  2 +-
 net/core/filter.c                             |  2 +-
 net/ipv4/ip_tunnel.c                          |  2 +-
 net/ipv6/icmp.c                               |  8 +++---
 net/ipv6/ila/ila_lwt.c                        |  4 +--
 net/ipv6/ip6_output.c                         | 18 ++++++------
 net/ipv6/ip6mr.c                              |  2 +-
 net/ipv6/ndisc.c                              |  2 +-
 net/ipv6/ping.c                               |  2 +-
 net/ipv6/raw.c                                |  4 +--
 net/ipv6/route.c                              | 28 +++++++++----------
 net/ipv6/tcp_ipv6.c                           |  4 +--
 net/ipv6/udp.c                                | 11 +++-----
 net/ipv6/xfrm6_policy.c                       |  2 +-
 net/l2tp/l2tp_ip6.c                           |  2 +-
 net/mpls/mpls_iptunnel.c                      |  2 +-
 net/netfilter/ipvs/ip_vs_xmit.c               | 14 +++++-----
 net/netfilter/nf_flow_table_core.c            |  8 ++----
 net/netfilter/nf_flow_table_ip.c              |  4 +--
 net/netfilter/nft_rt.c                        |  2 +-
 net/sctp/ipv6.c                               |  2 +-
 net/xfrm/xfrm_policy.c                        |  3 +-
 30 files changed, 77 insertions(+), 86 deletions(-)

drivers/infiniband/core/addr.c +3 -3

index f253295795f0a..f20dfe70fa0e4 100644
@@ -348,15 +348,15 @@ static int dst_fetch_ha(const struct dst_entry *dst,
 
 static bool has_gateway(const struct dst_entry *dst, sa_family_t family)
 {
-	struct rtable *rt;
-	struct rt6_info *rt6;
+	const struct rtable *rt;
+	const struct rt6_info *rt6;
 
 	if (family == AF_INET) {
 		rt = container_of(dst, struct rtable, dst);
 		return rt->rt_uses_gateway;
 	}
 
-	rt6 = container_of(dst, struct rt6_info, dst);
+	rt6 = dst_rt6_info(dst);
 	return rt6->rt6i_flags & RTF_GATEWAY;
 }

drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +1 -1

index b3472fb946177..dcd198104141f 100644
@@ -539,7 +539,7 @@ mlxsw_sp_span_gretap6_route(const struct net_device *to_dev,
 	if (!dst || dst->error)
 		goto out;
 
-	rt6 = container_of(dst, struct rt6_info, dst);
+	rt6 = dst_rt6_info(dst);
 
 	dev = dst->dev;
 	*saddrp = fl6.saddr;

drivers/net/vrf.c +1 -1

index 27761a884dc62..5968a3ab81770 100644
@@ -655,7 +655,7 @@ static int vrf_finish_output6(struct net *net, struct sock *sk,
 	skb->dev = dev;
 
 	rcu_read_lock();
-	nexthop = rt6_nexthop((struct rt6_info *)dst, &ipv6_hdr(skb)->daddr);
+	nexthop = rt6_nexthop(dst_rt6_info(dst), &ipv6_hdr(skb)->daddr);
 	neigh = __ipv6_neigh_lookup_noref(dst->dev, nexthop);
 	if (unlikely(!neigh))
 		neigh = __neigh_create(&nd_tbl, nexthop, dst->dev, false);

drivers/net/vxlan/vxlan_core.c +1 -1

index c114c91b558bd..ee02a92338da1 100644
@@ -2629,7 +2629,7 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		}
 
 		if (!info) {
-			u32 rt6i_flags = ((struct rt6_info *)ndst)->rt6i_flags;
+			u32 rt6i_flags = dst_rt6_info(ndst)->rt6i_flags;
 
 			err = encap_bypass_if_local(skb, dev, vxlan, dst,
 						    dst_port, ifindex, vni,

drivers/s390/net/qeth_core.h +2 -2

index 613eab7297046..5f17a2a5d0e33 100644
@@ -956,7 +956,7 @@ static inline struct dst_entry *qeth_dst_check_rcu(struct sk_buff *skb,
 	struct dst_entry *dst = skb_dst(skb);
 	struct rt6_info *rt;
 
-	rt = (struct rt6_info *) dst;
+	rt = dst_rt6_info(dst);
 	if (dst) {
 		if (proto == htons(ETH_P_IPV6))
 			dst = dst_check(dst, rt6_get_cookie(rt));

@@ -978,7 +978,7 @@ static inline __be32 qeth_next_hop_v4_rcu(struct sk_buff *skb,
 static inline struct in6_addr *qeth_next_hop_v6_rcu(struct sk_buff *skb,
 						    struct dst_entry *dst)
 {
-	struct rt6_info *rt = (struct rt6_info *) dst;
+	struct rt6_info *rt = dst_rt6_info(dst);
 
 	if (rt && !ipv6_addr_any(&rt->rt6i_gateway))
 		return &rt->rt6i_gateway;

include/net/ip6_fib.h +4 -2

index 9ba6413fd2e3e..1121d614942c8 100644
@@ -237,9 +237,11 @@ struct fib6_result {
 	for (rt = (w)->leaf; rt;					\
 	     rt = rcu_dereference_protected(rt->fib6_next, 1))
 
-static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst)
+#define dst_rt6_info(_ptr) container_of_const(_ptr, struct rt6_info, dst)
+
+static inline struct inet6_dev *ip6_dst_idev(const struct dst_entry *dst)
 {
-	return ((struct rt6_info *)dst)->rt6i_idev;
+	return dst_rt6_info(dst)->rt6i_idev;
 }
 
 static inline bool fib6_requires_src(const struct fib6_info *rt)

include/net/ip6_route.h +5 -6

index 61cfc8891f820..392232fcd703c 100644
@@ -218,12 +218,11 @@ void rt6_uncached_list_del(struct rt6_info *rt);
 static inline const struct rt6_info *skb_rt6_info(const struct sk_buff *skb)
 {
 	const struct dst_entry *dst = skb_dst(skb);
-	const struct rt6_info *rt6 = NULL;
 
 	if (dst)
-		rt6 = container_of(dst, struct rt6_info, dst);
+		return dst_rt6_info(dst);
 
-	return rt6;
+	return NULL;
 }
 
 /*

@@ -235,7 +234,7 @@ static inline void ip6_dst_store(struct sock *sk, struct dst_entry *dst,
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 
-	np->dst_cookie = rt6_get_cookie((struct rt6_info *)dst);
+	np->dst_cookie = rt6_get_cookie(dst_rt6_info(dst));
 	sk_setup_caps(sk, dst);
 	np->daddr_cache = daddr;
 #ifdef CONFIG_IPV6_SUBTREES

@@ -248,7 +247,7 @@ void ip6_sk_dst_store_flow(struct sock *sk, struct dst_entry *dst,
 
 static inline bool ipv6_unicast_destination(const struct sk_buff *skb)
 {
-	struct rt6_info *rt = (struct rt6_info *) skb_dst(skb);
+	const struct rt6_info *rt = dst_rt6_info(skb_dst(skb));
 
 	return rt->rt6i_flags & RTF_LOCAL;
 }

@@ -256,7 +255,7 @@ static inline bool ipv6_unicast_destination(const struct sk_buff *skb)
 static inline bool ipv6_anycast_destination(const struct dst_entry *dst,
 					    const struct in6_addr *daddr)
 {
-	struct rt6_info *rt = (struct rt6_info *)dst;
+	const struct rt6_info *rt = dst_rt6_info(dst);
 
 	return rt->rt6i_flags & RTF_ANYCAST ||
 		(rt->rt6i_dst.plen < 127 &&

net/bluetooth/6lowpan.c +1 -1

index 4eb1b3ced0d27..3bcc15c9415fc 100644
@@ -133,7 +133,7 @@ static inline struct lowpan_peer *peer_lookup_dst(struct lowpan_btle_dev *dev,
 						  struct in6_addr *daddr,
 						  struct sk_buff *skb)
 {
-	struct rt6_info *rt = (struct rt6_info *)skb_dst(skb);
+	struct rt6_info *rt = dst_rt6_info(skb_dst(skb));
 	int count = atomic_read(&dev->peer_count);
 	const struct in6_addr *nexthop;
 	struct lowpan_peer *peer;

net/core/dst_cache.c +1 -1

index 0ccfd5fa5cb9b..b17171345d649 100644
@@ -112,7 +112,7 @@ void dst_cache_set_ip6(struct dst_cache *dst_cache, struct dst_entry *dst,
 
 	idst = this_cpu_ptr(dst_cache->cache);
 	dst_cache_per_cpu_dst_set(this_cpu_ptr(dst_cache->cache), dst,
-				  rt6_get_cookie((struct rt6_info *)dst));
+				  rt6_get_cookie(dst_rt6_info(dst)));
 	idst->in6_saddr = *saddr;
 }
 EXPORT_SYMBOL_GPL(dst_cache_set_ip6);

net/core/filter.c +1 -1

index b64e7139eae19..bc52ab3374f3a 100644
@@ -2215,7 +2215,7 @@ static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb,
 	rcu_read_lock();
 	if (!nh) {
 		dst = skb_dst(skb);
-		nexthop = rt6_nexthop(container_of(dst, struct rt6_info, dst),
+		nexthop = rt6_nexthop(dst_rt6_info(dst),
 				      &ipv6_hdr(skb)->daddr);
 	} else {
 		nexthop = &nh->ipv6_nh;

net/ipv4/ip_tunnel.c +1 -1

index acf93f34a8213..72b2d68ef4da5 100644
@@ -544,7 +544,7 @@ static int tnl_update_pmtu(struct net_device *dev, struct sk_buff *skb,
 		struct rt6_info *rt6;
 		__be32 daddr;
 
-		rt6 = skb_valid_dst(skb) ? (struct rt6_info *)skb_dst(skb) :
+		rt6 = skb_valid_dst(skb) ? dst_rt6_info(skb_dst(skb)) :
 					   NULL;
 		daddr = md ? dst : tunnel->parms.iph.daddr;

net/ipv6/icmp.c +4 -4

index a790294d31048..25a3a726fa117 100644
@@ -214,7 +214,7 @@ static bool icmpv6_xrlim_allow(struct sock *sk, u8 type,
 	} else if (dst->dev && (dst->dev->flags&IFF_LOOPBACK)) {
 		res = true;
 	} else {
-		struct rt6_info *rt = (struct rt6_info *)dst;
+		struct rt6_info *rt = dst_rt6_info(dst);
 		int tmo = net->ipv6.sysctl.icmpv6_time;
 		struct inet_peer *peer;

@@ -245,7 +245,7 @@ static bool icmpv6_rt_has_prefsrc(struct sock *sk, u8 type,
 
 	dst = ip6_route_output(net, sk, fl6);
 	if (!dst->error) {
-		struct rt6_info *rt = (struct rt6_info *)dst;
+		struct rt6_info *rt = dst_rt6_info(dst);
 		struct in6_addr prefsrc;
 
 		rt6_get_prefsrc(rt, &prefsrc);

@@ -622,7 +622,7 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info,
 	if (ip6_append_data(sk, icmpv6_getfrag, &msg,
 			    len + sizeof(struct icmp6hdr),
 			    sizeof(struct icmp6hdr),
-			    &ipc6, &fl6, (struct rt6_info *)dst,
+			    &ipc6, &fl6, dst_rt6_info(dst),
 			    MSG_DONTWAIT)) {
 		ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTERRORS);
 		ip6_flush_pending_frames(sk);

@@ -811,7 +811,7 @@ static enum skb_drop_reason icmpv6_echo_reply(struct sk_buff *skb)
 	if (ip6_append_data(sk, icmpv6_getfrag, &msg,
 			    skb->len + sizeof(struct icmp6hdr),
 			    sizeof(struct icmp6hdr), &ipc6, &fl6,
-			    (struct rt6_info *)dst, MSG_DONTWAIT)) {
+			    dst_rt6_info(dst), MSG_DONTWAIT)) {
 		__ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTERRORS);
 		ip6_flush_pending_frames(sk);
 	} else {

net/ipv6/ila/ila_lwt.c +2 -2

index 9d37f7164e732..ff7e734e335b0 100644
@@ -38,7 +38,7 @@ static inline struct ila_params *ila_params_lwtunnel(
 static int ila_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	struct dst_entry *orig_dst = skb_dst(skb);
-	struct rt6_info *rt = (struct rt6_info *)orig_dst;
+	struct rt6_info *rt = dst_rt6_info(orig_dst);
 	struct ila_lwt *ilwt = ila_lwt_lwtunnel(orig_dst->lwtstate);
 	struct dst_entry *dst;
 	int err = -EINVAL;

@@ -72,7 +72,7 @@ static int ila_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 		memset(&fl6, 0, sizeof(fl6));
 		fl6.flowi6_oif = orig_dst->dev->ifindex;
 		fl6.flowi6_iif = LOOPBACK_IFINDEX;
-		fl6.daddr = *rt6_nexthop((struct rt6_info *)orig_dst,
+		fl6.daddr = *rt6_nexthop(dst_rt6_info(orig_dst),
 					 &ip6h->daddr);
 
 		dst = ip6_route_output(net, NULL, &fl6);

net/ipv6/ip6_output.c +9 -9

index 65e2f19814358..2341a4373bb94 100644
@@ -124,7 +124,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff *
 	IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len);
 
 	rcu_read_lock();
-	nexthop = rt6_nexthop((struct rt6_info *)dst, daddr);
+	nexthop = rt6_nexthop(dst_rt6_info(dst), daddr);
 	neigh = __ipv6_neigh_lookup_noref(dev, nexthop);
 
 	if (unlikely(IS_ERR_OR_NULL(neigh))) {

@@ -606,7 +606,7 @@ int ip6_forward(struct sk_buff *skb)
 		 *	send a redirect.
 		 */
 
-		rt = (struct rt6_info *) dst;
+		rt = dst_rt6_info(dst);
 		if (rt->rt6i_flags & RTF_GATEWAY)
 			target = &rt->rt6i_gateway;
 		else

@@ -861,7 +861,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
 		 int (*output)(struct net *, struct sock *, struct sk_buff *))
 {
 	struct sk_buff *frag;
-	struct rt6_info *rt = (struct rt6_info *)skb_dst(skb);
+	struct rt6_info *rt = dst_rt6_info(skb_dst(skb));
 	struct ipv6_pinfo *np = skb->sk && !dev_recursion_level() ?
 				inet6_sk(skb->sk) : NULL;
 	bool mono_delivery_time = skb->mono_delivery_time;

@@ -1069,7 +1069,7 @@ static struct dst_entry *ip6_sk_dst_check(struct sock *sk,
 		return NULL;
 	}
 
-	rt = (struct rt6_info *)dst;
+	rt = dst_rt6_info(dst);
 	/* Yes, checking route validity in not connected
 	 * case is not very simple. Take into account,
 	 * that we do not support routing by source, TOS,

@@ -1124,7 +1124,7 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk,
 		struct rt6_info *rt;
 
 		*dst = ip6_route_output(net, sk, fl6);
-		rt = (*dst)->error ? NULL : (struct rt6_info *)*dst;
+		rt = (*dst)->error ? NULL : dst_rt6_info(*dst);
 
 		rcu_read_lock();
 		from = rt ? rcu_dereference(rt->from) : NULL;

@@ -1166,7 +1166,7 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk,
 	 * dst entry and replace it instead with the
 	 * dst entry of the nexthop router
 	 */
-	rt = (struct rt6_info *) *dst;
+	rt = dst_rt6_info(*dst);
 	rcu_read_lock();
 	n = __ipv6_neigh_lookup_noref(rt->dst.dev,
 				      rt6_nexthop(rt, &fl6->daddr));

@@ -1500,7 +1500,7 @@ static int __ip6_append_data(struct sock *sk,
 	int offset = 0;
 	bool zc = false;
 	u32 tskey = 0;
-	struct rt6_info *rt = (struct rt6_info *)cork->dst;
+	struct rt6_info *rt = dst_rt6_info(cork->dst);
 	bool paged, hold_tskey, extra_uref = false;
 	struct ipv6_txoptions *opt = v6_cork->opt;
 	int csummode = CHECKSUM_NONE;

@@ -1959,7 +1959,7 @@ struct sk_buff *__ip6_make_skb(struct sock *sk,
 	struct net *net = sock_net(sk);
 	struct ipv6hdr *hdr;
 	struct ipv6_txoptions *opt = v6_cork->opt;
-	struct rt6_info *rt = (struct rt6_info *)cork->base.dst;
+	struct rt6_info *rt = dst_rt6_info(cork->base.dst);
 	struct flowi6 *fl6 = &cork->fl.u.ip6;
 	unsigned char proto = fl6->flowi6_proto;

@@ -2031,7 +2031,7 @@ struct sk_buff *__ip6_make_skb(struct sock *sk,
 int ip6_send_skb(struct sk_buff *skb)
 {
 	struct net *net = sock_net(skb->sk);
-	struct rt6_info *rt = (struct rt6_info *)skb_dst(skb);
+	struct rt6_info *rt = dst_rt6_info(skb_dst(skb));
 	int err;
 
 	rcu_read_lock();

net/ipv6/ip6mr.c +1 -1

index e24fa0843c7d1..1571e85a3531e 100644
@@ -2290,7 +2290,7 @@ int ip6mr_get_route(struct net *net, struct sk_buff *skb, struct rtmsg *rtm,
 	int err;
 	struct mr_table *mrt;
 	struct mfc6_cache *cache;
-	struct rt6_info *rt = (struct rt6_info *)skb_dst(skb);
+	struct rt6_info *rt = dst_rt6_info(skb_dst(skb));
 
 	rcu_read_lock();
 	mrt = __ip6mr_get_table(net, RT6_TABLE_DFLT);

net/ipv6/ndisc.c +1 -1

index 2062ab94721e3..23b46b5705c53 100644
@@ -1710,7 +1710,7 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target)
 	if (IS_ERR(dst))
 		return;
 
-	rt = (struct rt6_info *) dst;
+	rt = dst_rt6_info(dst);
 
 	if (rt->rt6i_flags & RTF_GATEWAY) {
 		ND_PRINTK(2, warn,

net/ipv6/ping.c +1 -1

index 25243737fbc42..878ed02b299d7 100644
@@ -154,7 +154,7 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	dst = ip6_sk_dst_lookup_flow(sk, &fl6, daddr, false);
 	if (IS_ERR(dst))
 		return PTR_ERR(dst);
-	rt = (struct rt6_info *) dst;
+	rt = dst_rt6_info(dst);
 
 	if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr))
 		fl6.flowi6_oif = np->mcast_oif;

net/ipv6/raw.c +2 -2

index 0a3e12502b05a..4f526606bc894 100644
@@ -592,7 +592,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
 	struct ipv6hdr *iph;
 	struct sk_buff *skb;
 	int err;
-	struct rt6_info *rt = (struct rt6_info *)*dstp;
+	struct rt6_info *rt = dst_rt6_info(*dstp);
 	int hlen = LL_RESERVED_SPACE(rt->dst.dev);
 	int tlen = rt->dst.dev->needed_tailroom;

@@ -911,7 +911,7 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		ipc6.opt = opt;
 		lock_sock(sk);
 		err = ip6_append_data(sk, raw6_getfrag, &rfv,
-			len, 0, &ipc6, &fl6, (struct rt6_info *)dst,
+			len, 0, &ipc6, &fl6, dst_rt6_info(dst),
 			msg->msg_flags);
 
 		if (err)

net/ipv6/route.c +14 ~~-14~~

index e320dfa7fe7fc..0fdd062d4b05b 100644
@@ -227,7 +227,7 @@ static struct neighbour *ip6_dst_neigh_lookup(const struct dst_entry *dst,
 					      struct sk_buff *skb,
 					      const void *daddr)
 {
-	const struct rt6_info *rt = container_of(dst, struct rt6_info, dst);
+	const struct rt6_info *rt = dst_rt6_info(dst);
 
 	return ip6_neigh_lookup(rt6_nexthop(rt, &in6addr_any),
 				dst->dev, skb, daddr);

@@ -235,8 +235,8 @@ static struct neighbour *ip6_dst_neigh_lookup(const struct dst_entry *dst,
 
 static void ip6_confirm_neigh(const struct dst_entry *dst, const void *daddr)
 {
+	const struct rt6_info *rt = dst_rt6_info(dst);
 	struct net_device *dev = dst->dev;
-	struct rt6_info *rt = (struct rt6_info *)dst;
 
 	daddr = choose_neigh_daddr(rt6_nexthop(rt, &in6addr_any), NULL, daddr);
 	if (!daddr)

@@ -355,7 +355,7 @@ EXPORT_SYMBOL(ip6_dst_alloc);
 
 static void ip6_dst_destroy(struct dst_entry *dst)
 {
-	struct rt6_info *rt = (struct rt6_info *)dst;
+	struct rt6_info *rt = dst_rt6_info(dst);
 	struct fib6_info *from;
 	struct inet6_dev *idev;

@@ -374,7 +374,7 @@ static void ip6_dst_destroy(struct dst_entry *dst)
 
 static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev)
 {
-	struct rt6_info *rt = (struct rt6_info *)dst;
+	struct rt6_info *rt = dst_rt6_info(dst);
 	struct inet6_dev *idev = rt->rt6i_idev;
 	struct fib6_info *from;

@@ -1284,7 +1284,7 @@ struct rt6_info *rt6_lookup(struct net *net, const struct in6_addr *daddr,
 
 	dst = fib6_rule_lookup(net, &fl6, skb, flags, ip6_pol_route_lookup);
 	if (dst->error == 0)
-		return (struct rt6_info *) dst;
+		return dst_rt6_info(dst);
 
 	dst_release(dst);

@@ -2641,7 +2641,7 @@ struct dst_entry *ip6_route_output_flags(struct net *net,
 
 	rcu_read_lock();
 	dst = ip6_route_output_flags_noref(net, sk, fl6, flags);
-	rt6 = (struct rt6_info *)dst;
+	rt6 = dst_rt6_info(dst);
 	/* For dst cached in uncached_list, refcnt is already taken. */
 	if (list_empty(&rt6->dst.rt_uncached) && !dst_hold_safe(dst)) {
 		dst = &net->ipv6.ip6_null_entry->dst;

@@ -2655,7 +2655,7 @@ EXPORT_SYMBOL_GPL(ip6_route_output_flags);
 
 struct dst_entry *ip6_blackhole_route(struct net *net, struct dst_entry *dst_orig)
 {
-	struct rt6_info *rt, *ort = (struct rt6_info *) dst_orig;
+	struct rt6_info *rt, *ort = dst_rt6_info(dst_orig);
 	struct net_device *loopback_dev = net->loopback_dev;
 	struct dst_entry *new = NULL;

@@ -2738,7 +2738,7 @@ INDIRECT_CALLABLE_SCOPE struct dst_entry *ip6_dst_check(struct dst_entry *dst,
 	struct fib6_info *from;
 	struct rt6_info *rt;
 
-	rt = container_of(dst, struct rt6_info, dst);
+	rt = dst_rt6_info(dst);
 
 	if (rt->sernum)
 		return rt6_is_valid(rt) ? dst : NULL;

@@ -2767,7 +2767,7 @@ EXPORT_INDIRECT_CALLABLE(ip6_dst_check);
 static void ip6_negative_advice(struct sock *sk,
 				struct dst_entry *dst)
 {
-	struct rt6_info *rt = (struct rt6_info *) dst;
+	struct rt6_info *rt = dst_rt6_info(dst);
 
 	if (rt->rt6i_flags & RTF_CACHE) {
 		rcu_read_lock();

@@ -2790,7 +2790,7 @@ static void ip6_link_failure(struct sk_buff *skb)
 
 	icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0);
 
-	rt = (struct rt6_info *) skb_dst(skb);
+	rt = dst_rt6_info(skb_dst(skb));
 	if (rt) {
 		rcu_read_lock();
 		if (rt->rt6i_flags & RTF_CACHE) {

@@ -2846,7 +2846,7 @@ static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk,
 				 bool confirm_neigh)
 {
 	const struct in6_addr *daddr, *saddr;
-	struct rt6_info *rt6 = (struct rt6_info *)dst;
+	struct rt6_info *rt6 = dst_rt6_info(dst);
 
 	/* Note: do *NOT* check dst_metric_locked(dst, RTAX_MTU)
 	 * IPv6 pmtu discovery isn't optional, so 'mtu lock' cannot disable it.

@@ -4169,7 +4169,7 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu
 		}
 	}
 
-	rt = (struct rt6_info *) dst;
+	rt = dst_rt6_info(dst);
 	if (rt->rt6i_flags & RTF_REJECT) {
 		net_dbg_ratelimited("rt6_redirect: source isn't a valid nexthop for redirect target\n");
 		return;

@@ -5604,7 +5604,7 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb,
 			 int iif, int type, u32 portid, u32 seq,
 			 unsigned int flags)
 {
-	struct rt6_info *rt6 = (struct rt6_info *)dst;
+	struct rt6_info *rt6 = dst_rt6_info(dst);
 	struct rt6key *rt6_dst, *rt6_src;
 	u32 *pmetrics, table, rt6_flags;
 	unsigned char nh_flags = 0;

@@ -6107,7 +6107,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh,
 	}
 
 
-	rt = container_of(dst, struct rt6_info, dst);
+	rt = dst_rt6_info(dst);
 	if (rt->dst.error) {
 		err = rt->dst.error;
 		ip6_rt_put(rt);

net/ipv6/tcp_ipv6.c +1 -3

index 83b48dc2b3ee2..64bdb6d978eed 100644
@@ -101,11 +101,9 @@ static void inet6_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb)
 	struct dst_entry *dst = skb_dst(skb);
 
 	if (dst && dst_hold_safe(dst)) {
-		const struct rt6_info *rt = (const struct rt6_info *)dst;
-
 		rcu_assign_pointer(sk->sk_rx_dst, dst);
 		sk->sk_rx_dst_ifindex = skb->skb_iif;
-		sk->sk_rx_dst_cookie = rt6_get_cookie(rt);
+		sk->sk_rx_dst_cookie = rt6_get_cookie(dst_rt6_info(dst));
 	}
 }

net/ipv6/udp.c +4 -7

index c77ee9a3cde24..954afe6ba883e 100644
@@ -913,11 +913,8 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
 
 static void udp6_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst)
 {
-	if (udp_sk_rx_dst_set(sk, dst)) {
-		const struct rt6_info *rt = (const struct rt6_info *)dst;
-
-		sk->sk_rx_dst_cookie = rt6_get_cookie(rt);
-	}
+	if (udp_sk_rx_dst_set(sk, dst))
+		sk->sk_rx_dst_cookie = rt6_get_cookie(dst_rt6_info(dst));
 }
 
 /* wrapper for udp_queue_rcv_skb tacking care of csum conversion and

@@ -1587,7 +1584,7 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 		skb = ip6_make_skb(sk, getfrag, msg, ulen,
 				   sizeof(struct udphdr), &ipc6,
-				   (struct rt6_info *)dst,
+				   dst_rt6_info(dst),
 				   msg->msg_flags, &cork);
 		err = PTR_ERR(skb);
 		if (!IS_ERR_OR_NULL(skb))

@@ -1614,7 +1611,7 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		ipc6.dontfrag = np->dontfrag;
 	up->len += ulen;
 	err = ip6_append_data(sk, getfrag, msg, ulen, sizeof(struct udphdr),
-			      &ipc6, fl6, (struct rt6_info *)dst,
+			      &ipc6, fl6, dst_rt6_info(dst),
 			      corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
 	if (err)
 		udp_v6_flush_pending_frames(sk);

net/ipv6/xfrm6_policy.c +1 -1

index 4cd625af91e6c..f24faa78ee82d 100644
@@ -77,7 +77,7 @@ static int xfrm6_get_saddr(xfrm_address_t *saddr,
 static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev,
 			  const struct flowi *fl)
 {
-	struct rt6_info *rt = (struct rt6_info *)xdst->route;
+	struct rt6_info *rt = dst_rt6_info(xdst->route);
 
 	xdst->u.dst.dev = dev;
 	netdev_hold(dev, &xdst->u.dst.dev_tracker, GFP_ATOMIC);

net/l2tp/l2tp_ip6.c +1 -1

index db4971d52802b..1b7a7b4f7273a 100644
@@ -630,7 +630,7 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	ulen = len + (skb_queue_empty(&sk->sk_write_queue) ? transhdrlen : 0);
 	err = ip6_append_data(sk, ip_generic_getfrag, msg,
 			      ulen, transhdrlen, &ipc6,
-			      &fl6, (struct rt6_info *)dst,
+			      &fl6, dst_rt6_info(dst),
 			      msg->msg_flags);
 	if (err)
 		ip6_flush_pending_frames(sk);

net/mpls/mpls_iptunnel.c +1 -1

index ef59e25dc4827..8985abcb7a058 100644
@@ -92,7 +92,7 @@ static int mpls_xmit(struct sk_buff *skb)
 			ttl = net->mpls.default_ttl;
 		else
 			ttl = ipv6_hdr(skb)->hop_limit;
-		rt6 = (struct rt6_info *)dst;
+		rt6 = dst_rt6_info(dst);
 	} else {
 		goto drop;
 	}

net/netfilter/ipvs/ip_vs_xmit.c +7 -7

index 65e0259178da4..5cd511162bc03 100644
@@ -180,7 +180,7 @@ static inline bool crosses_local_route_boundary(int skb_af, struct sk_buff *skb,
 			(!skb->dev || skb->dev->flags & IFF_LOOPBACK) &&
 			(addr_type & IPV6_ADDR_LOOPBACK);
 		old_rt_is_local = __ip_vs_is_local_route6(
-			(struct rt6_info *)skb_dst(skb));
+			dst_rt6_info(skb_dst(skb)));
 	} else
 #endif
 	{

@@ -481,7 +481,7 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb,
 	if (dest) {
 		dest_dst = __ip_vs_dst_check(dest);
 		if (likely(dest_dst))
-			rt = (struct rt6_info *) dest_dst->dst_cache;
+			rt = dst_rt6_info(dest_dst->dst_cache);
 		else {
 			u32 cookie;

@@ -501,7 +501,7 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb,
 				ip_vs_dest_dst_free(dest_dst);
 				goto err_unreach;
 			}
-			rt = (struct rt6_info *) dst;
+			rt = dst_rt6_info(dst);
 			cookie = rt6_get_cookie(rt);
 			__ip_vs_dst_set(dest, dest_dst, &rt->dst, cookie);
 			spin_unlock_bh(&dest->dst_lock);

@@ -517,7 +517,7 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb,
 					      rt_mode);
 		if (!dst)
 			goto err_unreach;
-		rt = (struct rt6_info *) dst;
+		rt = dst_rt6_info(dst);
 	}
 
 	local = __ip_vs_is_local_route6(rt);

@@ -862,7 +862,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 				      IP_VS_RT_MODE_RDR);
 	if (local < 0)
 		goto tx_error;
-	rt = (struct rt6_info *) skb_dst(skb);
+	rt = dst_rt6_info(skb_dst(skb));
 	/*
 	 * Avoid duplicate tuple in reply direction for NAT traffic
 	 * to local address when connection is sync-ed

@@ -1288,7 +1288,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	if (local)
 		return ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 1);
 
-	rt = (struct rt6_info *) skb_dst(skb);
+	rt = dst_rt6_info(skb_dst(skb));
 	tdev = rt->dst.dev;
 
 	/*

@@ -1590,7 +1590,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 				      &cp->daddr.in6, NULL, ipvsh, 0, rt_mode);
 	if (local < 0)
 		goto tx_error;
-	rt = (struct rt6_info *) skb_dst(skb);
+	rt = dst_rt6_info(skb_dst(skb));
 	/*
 	 * Avoid duplicate tuple in reply direction for NAT traffic
 	 * to local address when connection is sync-ed

net/netfilter/nf_flow_table_core.c +2 -6

index a0571339239c4..5c1ff07eaee0b 100644
@@ -77,12 +77,8 @@ EXPORT_SYMBOL_GPL(flow_offload_alloc);
 
 static u32 flow_offload_dst_cookie(struct flow_offload_tuple *flow_tuple)
 {
-	const struct rt6_info *rt;
-
-	if (flow_tuple->l3proto == NFPROTO_IPV6) {
-		rt = (const struct rt6_info *)flow_tuple->dst_cache;
-		return rt6_get_cookie(rt);
-	}
+	if (flow_tuple->l3proto == NFPROTO_IPV6)
+		return rt6_get_cookie(dst_rt6_info(flow_tuple->dst_cache));
 
 	return 0;
 }

net/netfilter/nf_flow_table_ip.c +2 -2

index 846fa2ad7c858..20dd4eeac78b1 100644
@@ -732,7 +732,7 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb,
 		return NF_ACCEPT;
 
 	if (unlikely(tuplehash->tuple.xmit_type == FLOW_OFFLOAD_XMIT_XFRM)) {
-		rt = (struct rt6_info *)tuplehash->tuple.dst_cache;
+		rt = dst_rt6_info(tuplehash->tuple.dst_cache);
 		memset(skb->cb, 0, sizeof(struct inet6_skb_parm));
 		IP6CB(skb)->iif = skb->dev->ifindex;
 		IP6CB(skb)->flags = IP6SKB_FORWARDED;

@@ -744,7 +744,7 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb,
 
 	switch (tuplehash->tuple.xmit_type) {
 	case FLOW_OFFLOAD_XMIT_NEIGH:
-		rt = (struct rt6_info *)tuplehash->tuple.dst_cache;
+		rt = dst_rt6_info(tuplehash->tuple.dst_cache);
 		outdev = rt->dst.dev;
 		skb->dev = outdev;
 		nexthop = rt6_nexthop(rt, &flow->tuplehash[!dir].tuple.src_v6);

net/netfilter/nft_rt.c +1 -1

index 24d9771385729..2434c624aafde 100644
@@ -80,7 +80,7 @@ void nft_rt_get_eval(const struct nft_expr *expr,
 		if (nft_pf(pkt) != NFPROTO_IPV6)
 			goto err;
 
-		memcpy(dest, rt6_nexthop((struct rt6_info *)dst,
+		memcpy(dest, rt6_nexthop(dst_rt6_info(dst),
 					 &ipv6_hdr(skb)->daddr),
 		       sizeof(struct in6_addr));
 		break;

net/sctp/ipv6.c +1 -1

index 08acda9ecdf56..717828e531621 100644
@@ -416,7 +416,7 @@ static void sctp_v6_get_dst(struct sctp_transport *t, union sctp_addr *saddr,
 	if (!IS_ERR_OR_NULL(dst)) {
 		struct rt6_info *rt;
 
-		rt = (struct rt6_info *)dst;
+		rt = dst_rt6_info(dst);
 		t->dst_cookie = rt6_get_cookie(rt);
 		pr_debug("rt6_dst:%pI6/%d rt6_src:%pI6\n",
 			 &rt->rt6i_dst.addr, rt->rt6i_dst.plen,

net/xfrm/xfrm_policy.c +1 -2

index d788baffbf104..68b3f9e7edffd 100644
@@ -2607,8 +2607,7 @@ static void xfrm_init_path(struct xfrm_dst *path, struct dst_entry *dst,
 			   int nfheader_len)
 {
 	if (dst->ops->family == AF_INET6) {
-		struct rt6_info *rt = (struct rt6_info *)dst;
-		path->path_cookie = rt6_get_cookie(rt);
+		path->path_cookie = rt6_get_cookie(dst_rt6_info(dst));
 		path->u.rt6.rt6i_nfheader_len = nfheader_len;
 	}
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #31 lore raw

PATCH 31/356 net/ipv6: release expired exception dst cached in socket

Link https://lore.kernel.org/netdev/20241113105611.GA6723@incl/T/#u open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiri Wiesner <[email protected]>

[ Upstream commit 3301ab7d5aeb0fe270f73a3d4810c9d1b6a9f045 ]

Dst objects get leaked in ip6_negative_advice() when this function is executed for an expired IPv6 route located in the exception table. There are several conditions that must be fulfilled for the leak to occur:
* an ICMPv6 packet indicating a change of the MTU for the path is received, resulting in an exception dst being created
* a TCP connection that uses the exception dst for routing packets must start timing out so that TCP begins retransmissions
* after the exception dst expires, the FIB6 garbage collector must not run before TCP executes ip6_negative_advice() for the expired exception dst

When TCP executes ip6_negative_advice() for an exception dst that has expired and if no other socket holds a reference to the exception dst, the refcount of the exception dst is 2, which corresponds to the increment made by dst_init() and the increment made by the TCP socket for which the connection is timing out. The refcount made by the socket is never released. The refcount of the dst is decremented in sk_dst_reset() but that decrement is counteracted by a dst_hold() intentionally placed just before the sk_dst_reset() in ip6_negative_advice(). After ip6_negative_advice() has finished, there is no other object tied to the dst. The socket lost its reference stored in sk_dst_cache and the dst is no longer in the exception table. The exception dst becomes a leaked object.

As a result of this dst leak, an unbalanced refcount is reported for the loopback device of a net namespace being destroyed under kernels that do not contain e5f80fcf869a ("ipv6: give an IPv6 dev to blackhole_netdev"):
unregister_netdevice: waiting for lo to become free. Usage count = 2

Fix the dst leak by removing the dst_hold() in ip6_negative_advice(). The patch that introduced the dst_hold() in ip6_negative_advice() was 92f1655aa2b22 ("net: fix __dst_negative_advice() race"). But 92f1655aa2b22 merely refactored the code with regards to the dst refcount so the issue was present even before 92f1655aa2b22. The bug was introduced in 54c1a859efd9f ("ipv6: Don't drop cache route entry unless timer actually expired.") where the expired cached route is deleted and the sk_dst_cache member of the socket is set to NULL by calling dst_negative_advice() but the refcount belonging to the socket is left unbalanced.

The IPv4 version - ipv4_negative_advice() - is not affected by this bug. When the TCP connection times out ipv4_negative_advice() merely resets the sk_dst_cache of the socket while decrementing the refcount of the exception dst.

Fixes 92f1655aa2b22 ("net: fix __dst_negative_advice() race")

Fixes 54c1a859efd9f ("ipv6: Don't drop cache route entry unless timer actually expired.")

Signed-off-by Jiri Wiesner <[email protected]>

Reviewed-by Eric Dumazet <[email protected]>

Link https://patch.msgid.link/20241128085950.GA4505@incl open here

Signed-off-by Sasha Levin <[email protected]>

---
 net/ipv6/route.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

net/ipv6/route.c +3 -3

index 0fdd062d4b05b..fc5c534620253 100644
@@ -2772,10 +2772,10 @@ static void ip6_negative_advice(struct sock *sk,
 	if (rt->rt6i_flags & RTF_CACHE) {
 		rcu_read_lock();
 		if (rt6_check_expired(rt)) {
-			/* counteract the dst_release() in sk_dst_reset() */
-			dst_hold(dst);
+			/* rt/dst can not be destroyed yet,
+			 * because of rcu_read_lock()
+			 */
 			sk_dst_reset(sk);
-
 			rt6_remove_exception_rt(rt);
 		}
 		rcu_read_unlock();

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #32 lore raw

PATCH 32/356 dccp: Fix memory leak in dccp_feat_change_recv

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ivan Solodovnikov <[email protected]>

[ Upstream commit 22be4727a8f898442066bcac34f8a1ad0bc72e14 ]

If dccp_feat_push_confirm() fails after new value for SP feature was accepted without reconciliation ('entry == NULL' branch), memory allocated for that value with dccp_feat_clone_sp_val() is never freed.

Here is the kmemleak stack for this:

unreferenced object 0xffff88801d4ab488 (size 8):
comm "syz-executor310", pid 1127, jiffies 4295085598 (age 41.666s) hex dump (first 8 bytes):

    01 b4 4a 1d 80 88 ff ff                          ..J.....
  backtrace:
    [<00000000db7cabfe>] kmemdup+0x23/0x50 mm/util.c:128
    [<0000000019b38405>] kmemdup include/linux/string.h:465 [inline]
    [<0000000019b38405>] dccp_feat_clone_sp_val net/dccp/feat.c:371 [inline]
    [<0000000019b38405>] dccp_feat_clone_sp_val net/dccp/feat.c:367 [inline]
    [<0000000019b38405>] dccp_feat_change_recv net/dccp/feat.c:1145 [inline]
    [<0000000019b38405>] dccp_feat_parse_options+0x1196/0x2180 net/dccp/feat.c:1416
    [<00000000b1f6d94a>] dccp_parse_options+0xa2a/0x1260 net/dccp/options.c:125
    [<0000000030d7b621>] dccp_rcv_state_process+0x197/0x13d0 net/dccp/input.c:650
    [<000000001f74c72e>] dccp_v4_do_rcv+0xf9/0x1a0 net/dccp/ipv4.c:688
    [<00000000a6c24128>] sk_backlog_rcv include/net/sock.h:1041 [inline]
    [<00000000a6c24128>] __release_sock+0x139/0x3b0 net/core/sock.c:2570
    [<00000000cf1f3a53>] release_sock+0x54/0x1b0 net/core/sock.c:3111
    [<000000008422fa23>] inet_wait_for_connect net/ipv4/af_inet.c:603 [inline]
    [<000000008422fa23>] __inet_stream_connect+0x5d0/0xf70 net/ipv4/af_inet.c:696
    [<0000000015b6f64d>] inet_stream_connect+0x53/0xa0 net/ipv4/af_inet.c:735
    [<0000000010122488>] __sys_connect_file+0x15c/0x1a0 net/socket.c:1865
    [<00000000b4b70023>] __sys_connect+0x165/0x1a0 net/socket.c:1882
    [<00000000f4cb3815>] __do_sys_connect net/socket.c:1892 [inline]
    [<00000000f4cb3815>] __se_sys_connect net/socket.c:1889 [inline]
    [<00000000f4cb3815>] __x64_sys_connect+0x6e/0xb0 net/socket.c:1889
    [<00000000e7b1e839>] do_syscall_64+0x33/0x40 arch/x86/entry/common.c:46
    [<0000000055e91434>] entry_SYSCALL_64_after_hwframe+0x67/0xd1

Clean up the allocated memory in case of dccp_feat_push_confirm() failure and bail out with an error reset code.

Found by Linux Verification Center (linuxtesting.org) with Syzkaller.

Fixes e77b8363b2ea ("dccp: Process incoming Change feature-negotiation options")

Signed-off-by Ivan Solodovnikov <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/dccp/feat.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

net/dccp/feat.c +5 -1

index 54086bb05c42c..f7554dcdaaba9 100644
@@ -1166,8 +1166,12 @@ static u8 dccp_feat_change_recv(struct list_head *fn, u8 is_mandatory, u8 opt,
 			goto not_valid_or_not_known;
 		}
 
-		return dccp_feat_push_confirm(fn, feat, local, &fval);
+		if (dccp_feat_push_confirm(fn, feat, local, &fval)) {
+			kfree(fval.sp.vec);
+			return DCCP_RESET_CODE_TOO_BUSY;
+		}
 
+		return 0;
 	} else if (entry->state == FEAT_UNSTABLE) {	/* 6.6.2 */
 		return 0;
 	}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #33 lore raw

PATCH 33/356 tipc: Fix use-after-free of kernel socket in cleanup_bearer().

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 6a2fa13312e51a621f652d522d7e2df7066330b6 ]

syzkaller reported a use-after-free of UDP kernel socket in cleanup_bearer() without repro. [0][1]

When bearer_disable() calls tipc_udp_disable(), cleanup of the UDP kernel socket is deferred by work calling cleanup_bearer().

tipc_net_stop() waits for such works to finish by checking tipc_net(net)->wq_count. However, the work decrements the count too early before releasing the kernel socket, unblocking cleanup_net() and resulting in use-after-free.

Let's move the decrement after releasing the socket in cleanup_bearer().

[0]:
ref_tracker: net notrefcnt@000000009b3d1faf has 1/1 users at

     sk_alloc+0x438/0x608
     inet_create+0x4c8/0xcb0
     __sock_create+0x350/0x6b8
     sock_create_kern+0x58/0x78
     udp_sock_create4+0x68/0x398
     udp_sock_create+0x88/0xc8
     tipc_udp_enable+0x5e8/0x848
     __tipc_nl_bearer_enable+0x84c/0xed8
     tipc_nl_bearer_enable+0x38/0x60
     genl_family_rcv_msg_doit+0x170/0x248
     genl_rcv_msg+0x400/0x5b0
     netlink_rcv_skb+0x1dc/0x398
     genl_rcv+0x44/0x68
     netlink_unicast+0x678/0x8b0
     netlink_sendmsg+0x5e4/0x898
     ____sys_sendmsg+0x500/0x830

[1]:

BUG: KMSAN: use-after-free in udp_hashslot include/net/udp.h:85 [inline]
BUG: KMSAN: use-after-free in udp_lib_unhash+0x3b8/0x930 net/ipv4/udp.c:1979
 udp_hashslot include/net/udp.h:85 [inline]
 udp_lib_unhash+0x3b8/0x930 net/ipv4/udp.c:1979
 sk_common_release+0xaf/0x3f0 net/core/sock.c:3820
 inet_release+0x1e0/0x260 net/ipv4/af_inet.c:437
 inet6_release+0x6f/0xd0 net/ipv6/af_inet6.c:489
 __sock_release net/socket.c:658 [inline]
 sock_release+0xa0/0x210 net/socket.c:686
 cleanup_bearer+0x42d/0x4c0 net/tipc/udp_media.c:819
 process_one_work kernel/workqueue.c:3229 [inline]
 process_scheduled_works+0xcaf/0x1c90 kernel/workqueue.c:3310
 worker_thread+0xf6c/0x1510 kernel/workqueue.c:3391
 kthread+0x531/0x6b0 kernel/kthread.c:389
 ret_from_fork+0x60/0x80 arch/x86/kernel/process.c:147
 ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:244

Uninit was created at:

 slab_free_hook mm/slub.c:2269 [inline]
 slab_free mm/slub.c:4580 [inline]
 kmem_cache_free+0x207/0xc40 mm/slub.c:4682
 net_free net/core/net_namespace.c:454 [inline]
 cleanup_net+0x16f2/0x19d0 net/core/net_namespace.c:647
 process_one_work kernel/workqueue.c:3229 [inline]
 process_scheduled_works+0xcaf/0x1c90 kernel/workqueue.c:3310
 worker_thread+0xf6c/0x1510 kernel/workqueue.c:3391
 kthread+0x531/0x6b0 kernel/kthread.c:389
 ret_from_fork+0x60/0x80 arch/x86/kernel/process.c:147
 ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:244

CPU: 0 UID: 0 PID: 54 Comm: kworker/0:2 Not tainted 6.12.0-rc1-00131-gf66ebf37d69c #7 91723d6f74857f70725e1583cba3cf4adc716cfa
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.3-0-ga6ed6b701f0a-prebuilt.qemu.org 04/01/2014
Workqueue: events cleanup_bearer

Fixes 26abe14379f8 ("net: Modify sk_alloc to not reference count the netns of kernel sockets.")

Reported-by syzkaller <[email protected]>

Signed-off-by Kuniyuki Iwashima <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/tipc/udp_media.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/tipc/udp_media.c +1 -1

index cdc8378261ec3..70a39e29a6352 100644
@@ -814,10 +814,10 @@ static void cleanup_bearer(struct work_struct *work)
 		kfree_rcu(rcast, rcu);
 	}
 
-	atomic_dec(&tipc_net(sock_net(ub->ubsock->sk))->wq_count);
 	dst_cache_destroy(&ub->rcast.dst_cache);
 	udp_tunnel_sock_release(ub->ubsock);
 	synchronize_net();
+	atomic_dec(&tipc_net(sock_net(ub->ubsock->sk))->wq_count);
 	kfree(ub);
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #34 lore raw

PATCH 34/356 net/smc: rename some fce to fce_v2x for clarity

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit ac053a169c71ceb0f25f784fce9ea720455097b4 ]

Rename some functions or variables with 'fce' in their name but used in SMCv2.1 as 'fce_v2x' for clarity.

Signed-off-by Wen Gu <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/smc_clc.c | 30 ++++++++++++++++--------------
 1 file changed, 16 insertions(+), 14 deletions(-)

net/smc/smc_clc.c +16 ~~-14~~

index 1489a8421d786..b34aff73ada4c 100644
@@ -428,15 +428,16 @@ smc_clc_msg_decl_valid(struct smc_clc_msg_decline *dclc)
 	return true;
 }
 
-static int smc_clc_fill_fce(struct smc_clc_first_contact_ext_v2x *fce,
-			    struct smc_init_info *ini)
+static int smc_clc_fill_fce_v2x(struct smc_clc_first_contact_ext_v2x *fce_v2x,
+				struct smc_init_info *ini)
 {
-	int ret = sizeof(*fce);
+	int ret = sizeof(*fce_v2x);
 
-	memset(fce, 0, sizeof(*fce));
-	fce->fce_v2_base.os_type = SMC_CLC_OS_LINUX;
-	fce->fce_v2_base.release = ini->release_nr;
-	memcpy(fce->fce_v2_base.hostname, smc_hostname, sizeof(smc_hostname));
+	memset(fce_v2x, 0, sizeof(*fce_v2x));
+	fce_v2x->fce_v2_base.os_type = SMC_CLC_OS_LINUX;
+	fce_v2x->fce_v2_base.release = ini->release_nr;
+	memcpy(fce_v2x->fce_v2_base.hostname,
+	       smc_hostname, sizeof(smc_hostname));
 	if (ini->is_smcd && ini->release_nr < SMC_RELEASE_1) {
 		ret = sizeof(struct smc_clc_first_contact_ext);
 		goto out;

@@ -444,8 +445,8 @@ static int smc_clc_fill_fce(struct smc_clc_first_contact_ext_v2x *fce,
 
 	if (ini->release_nr >= SMC_RELEASE_1) {
 		if (!ini->is_smcd) {
-			fce->max_conns = ini->max_conns;
-			fce->max_links = ini->max_links;
+			fce_v2x->max_conns = ini->max_conns;
+			fce_v2x->max_links = ini->max_links;
 		}
 	}

@@ -1012,8 +1013,8 @@ static int smc_clc_send_confirm_accept(struct smc_sock *smc,
 				       int first_contact, u8 version,
 				       u8 *eid, struct smc_init_info *ini)
 {
+	struct smc_clc_first_contact_ext_v2x fce_v2x;
 	struct smc_connection *conn = &smc->conn;
-	struct smc_clc_first_contact_ext_v2x fce;
 	struct smcd_dev *smcd = conn->lgr->smcd;
 	struct smc_clc_msg_accept_confirm *clc;
 	struct smc_clc_fce_gid_ext gle;

@@ -1045,7 +1046,7 @@ static int smc_clc_send_confirm_accept(struct smc_sock *smc,
 				memcpy(clc_v2->d1.eid, eid, SMC_MAX_EID_LEN);
 			len = SMCD_CLC_ACCEPT_CONFIRM_LEN_V2;
 			if (first_contact) {
-				fce_len = smc_clc_fill_fce(&fce, ini);
+				fce_len = smc_clc_fill_fce_v2x(&fce_v2x, ini);
 				len += fce_len;
 			}
 			clc_v2->hdr.length = htons(len);

@@ -1091,9 +1092,10 @@ static int smc_clc_send_confirm_accept(struct smc_sock *smc,
 				memcpy(clc_v2->r1.eid, eid, SMC_MAX_EID_LEN);
 			len = SMCR_CLC_ACCEPT_CONFIRM_LEN_V2;
 			if (first_contact) {
-				fce_len = smc_clc_fill_fce(&fce, ini);
+				fce_len = smc_clc_fill_fce_v2x(&fce_v2x, ini);
 				len += fce_len;
-				fce.fce_v2_base.v2_direct = !link->lgr->uses_gateway;
+				fce_v2x.fce_v2_base.v2_direct =
+					!link->lgr->uses_gateway;
 				if (clc->hdr.type == SMC_CLC_CONFIRM) {
 					memset(&gle, 0, sizeof(gle));
 					gle.gid_cnt = ini->smcrv2.gidlist.len;

@@ -1120,7 +1122,7 @@ static int smc_clc_send_confirm_accept(struct smc_sock *smc,
 						SMCR_CLC_ACCEPT_CONFIRM_LEN) -
 				   sizeof(trl);
 	if (version > SMC_V1 && first_contact) {
-		vec[i].iov_base = &fce;
+		vec[i].iov_base = &fce_v2x;
 		vec[i++].iov_len = fce_len;
 		if (!conn->lgr->is_smcd) {
 			if (clc->hdr.type == SMC_CLC_CONFIRM) {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #35 lore raw

PATCH 35/356 net/smc: introduce sub-functions for smc_clc_send_confirm_accept()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit 5205ac4483b630e47c65f192a3ac19be7a8ea648 ]

There is a large if-else block in smc_clc_send_confirm_accept() and it is better to split it into two sub-functions.

Signed-off-by Wen Gu <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/smc_clc.c | 197 +++++++++++++++++++++++++++-------------------
 1 file changed, 115 insertions(+), 82 deletions(-)

net/smc/smc_clc.c +115 ~~-82~~

index b34aff73ada4c..d471a06baac32 100644
@@ -1007,6 +1007,112 @@ int smc_clc_send_proposal(struct smc_sock *smc, struct smc_init_info *ini)
 	return reason_code;
 }
 
+static void
+smcd_clc_prep_confirm_accept(struct smc_connection *conn,
+			     struct smc_clc_msg_accept_confirm_v2 *clc_v2,
+			     int first_contact, u8 version,
+			     u8 *eid, struct smc_init_info *ini,
+			     int *fce_len,
+			     struct smc_clc_first_contact_ext_v2x *fce_v2x,
+			     struct smc_clc_msg_trail *trl)
+{
+	struct smcd_dev *smcd = conn->lgr->smcd;
+	struct smc_clc_msg_accept_confirm *clc;
+	int len;
+
+	/* SMC-D specific settings */
+	clc = (struct smc_clc_msg_accept_confirm *)clc_v2;
+	memcpy(clc->hdr.eyecatcher, SMCD_EYECATCHER,
+	       sizeof(SMCD_EYECATCHER));
+	clc->hdr.typev1 = SMC_TYPE_D;
+	clc->d0.gid = htonll(smcd->ops->get_local_gid(smcd));
+	clc->d0.token = htonll(conn->rmb_desc->token);
+	clc->d0.dmbe_size = conn->rmbe_size_comp;
+	clc->d0.dmbe_idx = 0;
+	memcpy(&clc->d0.linkid, conn->lgr->id, SMC_LGR_ID_SIZE);
+	if (version == SMC_V1) {
+		clc->hdr.length = htons(SMCD_CLC_ACCEPT_CONFIRM_LEN);
+	} else {
+		clc_v2->d1.chid = htons(smc_ism_get_chid(smcd));
+		if (eid && eid[0])
+			memcpy(clc_v2->d1.eid, eid, SMC_MAX_EID_LEN);
+		len = SMCD_CLC_ACCEPT_CONFIRM_LEN_V2;
+		if (first_contact) {
+			*fce_len = smc_clc_fill_fce_v2x(fce_v2x, ini);
+			len += *fce_len;
+		}
+		clc_v2->hdr.length = htons(len);
+	}
+	memcpy(trl->eyecatcher, SMCD_EYECATCHER,
+	       sizeof(SMCD_EYECATCHER));
+}
+
+static void
+smcr_clc_prep_confirm_accept(struct smc_connection *conn,
+			     struct smc_clc_msg_accept_confirm_v2 *clc_v2,
+			     int first_contact, u8 version,
+			     u8 *eid, struct smc_init_info *ini,
+			     int *fce_len,
+			     struct smc_clc_first_contact_ext_v2x *fce_v2x,
+			     struct smc_clc_fce_gid_ext *gle,
+			     struct smc_clc_msg_trail *trl)
+{
+	struct smc_clc_msg_accept_confirm *clc;
+	struct smc_link *link = conn->lnk;
+	int len;
+
+	/* SMC-R specific settings */
+	clc = (struct smc_clc_msg_accept_confirm *)clc_v2;
+	memcpy(clc->hdr.eyecatcher, SMC_EYECATCHER,
+	       sizeof(SMC_EYECATCHER));
+	clc->hdr.typev1 = SMC_TYPE_R;
+	memcpy(clc->r0.lcl.id_for_peer, local_systemid,
+	       sizeof(local_systemid));
+	memcpy(&clc->r0.lcl.gid, link->gid, SMC_GID_SIZE);
+	memcpy(&clc->r0.lcl.mac, &link->smcibdev->mac[link->ibport - 1],
+	       ETH_ALEN);
+	hton24(clc->r0.qpn, link->roce_qp->qp_num);
+	clc->r0.rmb_rkey =
+		htonl(conn->rmb_desc->mr[link->link_idx]->rkey);
+	clc->r0.rmbe_idx = 1; /* for now: 1 RMB = 1 RMBE */
+	clc->r0.rmbe_alert_token = htonl(conn->alert_token_local);
+	switch (clc->hdr.type) {
+	case SMC_CLC_ACCEPT:
+		clc->r0.qp_mtu = link->path_mtu;
+		break;
+	case SMC_CLC_CONFIRM:
+		clc->r0.qp_mtu = min(link->path_mtu, link->peer_mtu);
+		break;
+	}
+	clc->r0.rmbe_size = conn->rmbe_size_comp;
+	clc->r0.rmb_dma_addr = conn->rmb_desc->is_vm ?
+		cpu_to_be64((uintptr_t)conn->rmb_desc->cpu_addr) :
+		cpu_to_be64((u64)sg_dma_address
+			    (conn->rmb_desc->sgt[link->link_idx].sgl));
+	hton24(clc->r0.psn, link->psn_initial);
+	if (version == SMC_V1) {
+		clc->hdr.length = htons(SMCR_CLC_ACCEPT_CONFIRM_LEN);
+	} else {
+		if (eid && eid[0])
+			memcpy(clc_v2->r1.eid, eid, SMC_MAX_EID_LEN);
+		len = SMCR_CLC_ACCEPT_CONFIRM_LEN_V2;
+		if (first_contact) {
+			*fce_len = smc_clc_fill_fce_v2x(fce_v2x, ini);
+			len += *fce_len;
+			fce_v2x->fce_v2_base.v2_direct =
+				!link->lgr->uses_gateway;
+			if (clc->hdr.type == SMC_CLC_CONFIRM) {
+				memset(gle, 0, sizeof(*gle));
+				gle->gid_cnt = ini->smcrv2.gidlist.len;
+				len += sizeof(*gle);
+				len += gle->gid_cnt * sizeof(gle->gid[0]);
+			}
+		}
+		clc_v2->hdr.length = htons(len);
+	}
+	memcpy(trl->eyecatcher, SMC_EYECATCHER, sizeof(SMC_EYECATCHER));
+}
+
 /* build and send CLC CONFIRM / ACCEPT message */
 static int smc_clc_send_confirm_accept(struct smc_sock *smc,
 				       struct smc_clc_msg_accept_confirm_v2 *clc_v2,

@@ -1015,11 +1121,10 @@ static int smc_clc_send_confirm_accept(struct smc_sock *smc,
 {
 	struct smc_clc_first_contact_ext_v2x fce_v2x;
 	struct smc_connection *conn = &smc->conn;
-	struct smcd_dev *smcd = conn->lgr->smcd;
 	struct smc_clc_msg_accept_confirm *clc;
 	struct smc_clc_fce_gid_ext gle;
 	struct smc_clc_msg_trail trl;
-	int i, len, fce_len;
+	int i, fce_len;
 	struct kvec vec[5];
 	struct msghdr msg;

@@ -1028,86 +1133,14 @@ static int smc_clc_send_confirm_accept(struct smc_sock *smc,
 	clc->hdr.version = version;	/* SMC version */
 	if (first_contact)
 		clc->hdr.typev2 |= SMC_FIRST_CONTACT_MASK;
-	if (conn->lgr->is_smcd) {
-		/* SMC-D specific settings */
-		memcpy(clc->hdr.eyecatcher, SMCD_EYECATCHER,
-		       sizeof(SMCD_EYECATCHER));
-		clc->hdr.typev1 = SMC_TYPE_D;
-		clc->d0.gid = htonll(smcd->ops->get_local_gid(smcd));
-		clc->d0.token = htonll(conn->rmb_desc->token);
-		clc->d0.dmbe_size = conn->rmbe_size_comp;
-		clc->d0.dmbe_idx = 0;
-		memcpy(&clc->d0.linkid, conn->lgr->id, SMC_LGR_ID_SIZE);
-		if (version == SMC_V1) {
-			clc->hdr.length = htons(SMCD_CLC_ACCEPT_CONFIRM_LEN);
-		} else {
-			clc_v2->d1.chid = htons(smc_ism_get_chid(smcd));
-			if (eid && eid[0])
-				memcpy(clc_v2->d1.eid, eid, SMC_MAX_EID_LEN);
-			len = SMCD_CLC_ACCEPT_CONFIRM_LEN_V2;
-			if (first_contact) {
-				fce_len = smc_clc_fill_fce_v2x(&fce_v2x, ini);
-				len += fce_len;
-			}
-			clc_v2->hdr.length = htons(len);
-		}
-		memcpy(trl.eyecatcher, SMCD_EYECATCHER,
-		       sizeof(SMCD_EYECATCHER));
-	} else {
-		struct smc_link *link = conn->lnk;
-
-		/* SMC-R specific settings */
-		memcpy(clc->hdr.eyecatcher, SMC_EYECATCHER,
-		       sizeof(SMC_EYECATCHER));
-		clc->hdr.typev1 = SMC_TYPE_R;
-		clc->hdr.length = htons(SMCR_CLC_ACCEPT_CONFIRM_LEN);
-		memcpy(clc->r0.lcl.id_for_peer, local_systemid,
-		       sizeof(local_systemid));
-		memcpy(&clc->r0.lcl.gid, link->gid, SMC_GID_SIZE);
-		memcpy(&clc->r0.lcl.mac, &link->smcibdev->mac[link->ibport - 1],
-		       ETH_ALEN);
-		hton24(clc->r0.qpn, link->roce_qp->qp_num);
-		clc->r0.rmb_rkey =
-			htonl(conn->rmb_desc->mr[link->link_idx]->rkey);
-		clc->r0.rmbe_idx = 1; /* for now: 1 RMB = 1 RMBE */
-		clc->r0.rmbe_alert_token = htonl(conn->alert_token_local);
-		switch (clc->hdr.type) {
-		case SMC_CLC_ACCEPT:
-			clc->r0.qp_mtu = link->path_mtu;
-			break;
-		case SMC_CLC_CONFIRM:
-			clc->r0.qp_mtu = min(link->path_mtu, link->peer_mtu);
-			break;
-		}
-		clc->r0.rmbe_size = conn->rmbe_size_comp;
-		clc->r0.rmb_dma_addr = conn->rmb_desc->is_vm ?
-			cpu_to_be64((uintptr_t)conn->rmb_desc->cpu_addr) :
-			cpu_to_be64((u64)sg_dma_address
-				    (conn->rmb_desc->sgt[link->link_idx].sgl));
-		hton24(clc->r0.psn, link->psn_initial);
-		if (version == SMC_V1) {
-			clc->hdr.length = htons(SMCR_CLC_ACCEPT_CONFIRM_LEN);
-		} else {
-			if (eid && eid[0])
-				memcpy(clc_v2->r1.eid, eid, SMC_MAX_EID_LEN);
-			len = SMCR_CLC_ACCEPT_CONFIRM_LEN_V2;
-			if (first_contact) {
-				fce_len = smc_clc_fill_fce_v2x(&fce_v2x, ini);
-				len += fce_len;
-				fce_v2x.fce_v2_base.v2_direct =
-					!link->lgr->uses_gateway;
-				if (clc->hdr.type == SMC_CLC_CONFIRM) {
-					memset(&gle, 0, sizeof(gle));
-					gle.gid_cnt = ini->smcrv2.gidlist.len;
-					len += sizeof(gle);
-					len += gle.gid_cnt * sizeof(gle.gid[0]);
-				}
-			}
-			clc_v2->hdr.length = htons(len);
-		}
-		memcpy(trl.eyecatcher, SMC_EYECATCHER, sizeof(SMC_EYECATCHER));
-	}
-
+	if (conn->lgr->is_smcd)
+		smcd_clc_prep_confirm_accept(conn, clc_v2, first_contact,
+					     version, eid, ini, &fce_len,
+					     &fce_v2x, &trl);
+	else
+		smcr_clc_prep_confirm_accept(conn, clc_v2, first_contact,
+					     version, eid, ini, &fce_len,
+					     &fce_v2x, &gle, &trl);
 	memset(&msg, 0, sizeof(msg));
 	i = 0;
 	vec[i].iov_base = clc_v2;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:31 UTC #36 lore raw

PATCH 36/356 net/smc: unify the structs of accept or confirm message for v1 and v2

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit 9505450d55b0f7809fe63c36ad9339a909461c87 ]

The structs of CLC accept and confirm messages for SMCv1 and SMCv2 are separately defined and often casted to each other in the code, which may increase the risk of errors caused by future divergence of them. So unify them into one struct for better maintainability.

Signed-off-by Wen Gu <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/af_smc.c  | 52 +++++++++++++------------------------
 net/smc/smc_clc.c | 65 ++++++++++++++++++++---------------------------
 net/smc/smc_clc.h | 42 +++++++++++++-----------------
 3 files changed, 62 insertions(+), 97 deletions(-)

net/smc/af_smc.c +17 ~~-35~~

index 3158b94fd347a..ea24fb4dae0df 100644
@@ -659,8 +659,6 @@ static bool smc_isascii(char *hostname)
 static void smc_conn_save_peer_info_fce(struct smc_sock *smc,
 					struct smc_clc_msg_accept_confirm *clc)
 {
-	struct smc_clc_msg_accept_confirm_v2 *clc_v2 =
-		(struct smc_clc_msg_accept_confirm_v2 *)clc;
 	struct smc_clc_first_contact_ext *fce;
 	int clc_v2_len;

@@ -669,17 +667,15 @@ static void smc_conn_save_peer_info_fce(struct smc_sock *smc,
 		return;
 
 	if (smc->conn.lgr->is_smcd) {
-		memcpy(smc->conn.lgr->negotiated_eid, clc_v2->d1.eid,
+		memcpy(smc->conn.lgr->negotiated_eid, clc->d1.eid,
 		       SMC_MAX_EID_LEN);
-		clc_v2_len = offsetofend(struct smc_clc_msg_accept_confirm_v2,
-					 d1);
+		clc_v2_len = offsetofend(struct smc_clc_msg_accept_confirm, d1);
 	} else {
-		memcpy(smc->conn.lgr->negotiated_eid, clc_v2->r1.eid,
+		memcpy(smc->conn.lgr->negotiated_eid, clc->r1.eid,
 		       SMC_MAX_EID_LEN);
-		clc_v2_len = offsetofend(struct smc_clc_msg_accept_confirm_v2,
-					 r1);
+		clc_v2_len = offsetofend(struct smc_clc_msg_accept_confirm, r1);
 	}
-	fce = (struct smc_clc_first_contact_ext *)(((u8 *)clc_v2) + clc_v2_len);
+	fce = (struct smc_clc_first_contact_ext *)(((u8 *)clc) + clc_v2_len);
 	smc->conn.lgr->peer_os = fce->os_type;
 	smc->conn.lgr->peer_smc_release = fce->release;
 	if (smc_isascii(fce->hostname))

@@ -1131,13 +1127,13 @@ static int smc_connect_ism_vlan_cleanup(struct smc_sock *smc,
 }
 
 #define SMC_CLC_MAX_ACCEPT_LEN \
-	(sizeof(struct smc_clc_msg_accept_confirm_v2) + \
+	(sizeof(struct smc_clc_msg_accept_confirm) + \
 	 sizeof(struct smc_clc_first_contact_ext_v2x) + \
 	 sizeof(struct smc_clc_msg_trail))
 
 /* CLC handshake during connect */
 static int smc_connect_clc(struct smc_sock *smc,
-			   struct smc_clc_msg_accept_confirm_v2 *aclc2,
+			   struct smc_clc_msg_accept_confirm *aclc,
 			   struct smc_init_info *ini)
 {
 	int rc = 0;

@@ -1147,7 +1143,7 @@ static int smc_connect_clc(struct smc_sock *smc,
 	if (rc)
 		return rc;
 	/* receive SMC Accept CLC message */
-	return smc_clc_wait_msg(smc, aclc2, SMC_CLC_MAX_ACCEPT_LEN,
+	return smc_clc_wait_msg(smc, aclc, SMC_CLC_MAX_ACCEPT_LEN,
 				SMC_CLC_ACCEPT, CLC_WAIT_TIME);
 }

@@ -1183,10 +1179,8 @@ static int smc_connect_rdma_v2_prepare(struct smc_sock *smc,
 				       struct smc_clc_msg_accept_confirm *aclc,
 				       struct smc_init_info *ini)
 {
-	struct smc_clc_msg_accept_confirm_v2 *clc_v2 =
-		(struct smc_clc_msg_accept_confirm_v2 *)aclc;
 	struct smc_clc_first_contact_ext *fce =
-		smc_get_clc_first_contact_ext(clc_v2, false);
+		smc_get_clc_first_contact_ext(aclc, false);
 	struct net *net = sock_net(&smc->sk);
 	int rc;

@@ -1309,10 +1303,7 @@ static int smc_connect_rdma(struct smc_sock *smc,
 	}
 
 	if (aclc->hdr.version > SMC_V1) {
-		struct smc_clc_msg_accept_confirm_v2 *clc_v2 =
-			(struct smc_clc_msg_accept_confirm_v2 *)aclc;
-
-		eid = clc_v2->r1.eid;
+		eid = aclc->r1.eid;
 		if (ini->first_contact_local)
 			smc_fill_gid_list(link->lgr, &ini->smcrv2.gidlist,
 					  link->smcibdev, link->gid);

@@ -1353,7 +1344,7 @@ static int smc_connect_rdma(struct smc_sock *smc,
  * Determine from the CHID of the received CLC ACCEPT the ISM device chosen.
  */
 static int
-smc_v2_determine_accepted_chid(struct smc_clc_msg_accept_confirm_v2 *aclc,
+smc_v2_determine_accepted_chid(struct smc_clc_msg_accept_confirm *aclc,
 			       struct smc_init_info *ini)
 {
 	int i;

@@ -1380,12 +1371,9 @@ static int smc_connect_ism(struct smc_sock *smc,
 	ini->first_contact_peer = aclc->hdr.typev2 & SMC_FIRST_CONTACT_MASK;
 
 	if (aclc->hdr.version == SMC_V2) {
-		struct smc_clc_msg_accept_confirm_v2 *aclc_v2 =
-			(struct smc_clc_msg_accept_confirm_v2 *)aclc;
-
 		if (ini->first_contact_peer) {
 			struct smc_clc_first_contact_ext *fce =
-				smc_get_clc_first_contact_ext(aclc_v2, true);
+				smc_get_clc_first_contact_ext(aclc, true);
 
 			ini->release_nr = fce->release;
 			rc = smc_clc_clnt_v2x_features_validate(fce, ini);

@@ -1393,7 +1381,7 @@ static int smc_connect_ism(struct smc_sock *smc,
 				return rc;
 		}
 
-		rc = smc_v2_determine_accepted_chid(aclc_v2, ini);
+		rc = smc_v2_determine_accepted_chid(aclc, ini);
 		if (rc)
 			return rc;
 	}

@@ -1419,12 +1407,8 @@ static int smc_connect_ism(struct smc_sock *smc,
 	smc_rx_init(smc);
 	smc_tx_init(smc);
 
-	if (aclc->hdr.version > SMC_V1) {
-		struct smc_clc_msg_accept_confirm_v2 *clc_v2 =
-			(struct smc_clc_msg_accept_confirm_v2 *)aclc;
-
-		eid = clc_v2->d1.eid;
-	}
+	if (aclc->hdr.version > SMC_V1)
+		eid = aclc->d1.eid;
 
 	rc = smc_clc_send_confirm(smc, ini->first_contact_local,
 				  aclc->hdr.version, eid, ini);

@@ -1475,7 +1459,6 @@ static int smc_connect_check_aclc(struct smc_init_info *ini,
 static int __smc_connect(struct smc_sock *smc)
 {
 	u8 version = smc_ism_is_v2_capable() ? SMC_V2 : SMC_V1;
-	struct smc_clc_msg_accept_confirm_v2 *aclc2;
 	struct smc_clc_msg_accept_confirm *aclc;
 	struct smc_init_info *ini = NULL;
 	u8 *buf = NULL;

@@ -1523,11 +1506,10 @@ static int __smc_connect(struct smc_sock *smc)
 		rc = SMC_CLC_DECL_MEM;
 		goto fallback;
 	}
-	aclc2 = (struct smc_clc_msg_accept_confirm_v2 *)buf;
-	aclc = (struct smc_clc_msg_accept_confirm *)aclc2;
+	aclc = (struct smc_clc_msg_accept_confirm *)buf;
 
 	/* perform CLC handshake */
-	rc = smc_connect_clc(smc, aclc2, ini);
+	rc = smc_connect_clc(smc, aclc, ini);
 	if (rc) {
 		/* -EAGAIN on timeout, see tcp_recvmsg() */
 		if (rc == -EAGAIN) {

net/smc/smc_clc.c +28 ~~-37~~

196 lines not on this page · raw message

net/smc/smc_clc.h +17 ~~-25~~

71 lines not on this page · raw message

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #37 lore raw

PATCH 37/356 net/smc: define a reserved CHID range for virtual ISM devices

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit 8dd512df3c98ce8081e3541990bf849157675723 ]

According to virtual ISM support feature defined by SMCv2.1, CHIDs in the range 0xFF00 to 0xFFFF are reserved for use by virtual ISM devices.

And two helpers are introduced to distinguish virtual ISM devices from the existing platform firmware ISM devices.

Signed-off-by Wen Gu <[email protected]>

Reviewed-and-tested-by Wenjia Zhang <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/smc_ism.h | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

net/smc/smc_ism.h +20 -0

index 832b2f42d79f3..d1228a615f23c 100644
@@ -15,6 +15,8 @@
 
 #include "smc.h"
 
+#define SMC_VIRTUAL_ISM_CHID_MASK	0xFF00
+
 struct smcd_dev_list {	/* List of SMCD devices */
 	struct list_head list;
 	struct mutex mutex;	/* Protects list of devices */

@@ -56,4 +58,22 @@ static inline int smc_ism_write(struct smcd_dev *smcd, u64 dmb_tok,
 	return rc < 0 ? rc : 0;
 }
 
+static inline bool __smc_ism_is_virtual(u16 chid)
+{
+	/* CHIDs in range of 0xFF00 to 0xFFFF are reserved
+	 * for virtual ISM device.
+	 *
+	 * loopback-ism:	0xFFFF
+	 * virtio-ism:		0xFF00 ~ 0xFFFE
+	 */
+	return ((chid & 0xFF00) == 0xFF00);
+}
+
+static inline bool smc_ism_is_virtual(struct smcd_dev *smcd)
+{
+	u16 chid = smcd->ops->get_chid(smcd);
+
+	return __smc_ism_is_virtual(chid);
+}
+
 #endif

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #38 lore raw

PATCH 38/356 net/smc: compatible with 128-bits extended GID of virtual ISM device

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit b40584d145700addc70cc29e4f0850a4ed955b1c ]

According to virtual ISM support feature defined by SMCv2.1, GIDs of virtual ISM device are UUIDs defined by RFC4122, which are 128-bits long. So some adaptation work is required. And note that the GIDs of existing platform firmware ISM devices still remain 64-bits long.

Signed-off-by Wen Gu <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/s390/net/ism_drv.c | 19 ++++++-----
 include/net/smc.h          | 15 ++++++---
 net/smc/af_smc.c           | 66 ++++++++++++++++++++++++++++++--------
 net/smc/smc.h              |  3 --
 net/smc/smc_clc.c          | 43 ++++++++++++++++++-------
 net/smc/smc_clc.h          | 10 ++++--
 net/smc/smc_core.c         | 31 ++++++++++++------
 net/smc/smc_core.h         | 17 +++++++---
 net/smc/smc_diag.c         |  7 ++--
 net/smc/smc_ism.c          | 17 ++++++----
 net/smc/smc_ism.h          |  3 +-
 net/smc/smc_pnet.c         |  4 +--
 12 files changed, 167 insertions(+), 68 deletions(-)

drivers/s390/net/ism_drv.c +11 -8

index 622a61f8a3b84..f6a0626a6b3ec 100644
@@ -762,10 +762,10 @@ static int ism_query_rgid(struct ism_dev *ism, u64 rgid, u32 vid_valid,
 	return ism_cmd(ism, &cmd);
 }
 
-static int smcd_query_rgid(struct smcd_dev *smcd, u64 rgid, u32 vid_valid,
-			   u32 vid)
+static int smcd_query_rgid(struct smcd_dev *smcd, struct smcd_gid *rgid,
+			   u32 vid_valid, u32 vid)
 {
-	return ism_query_rgid(smcd->priv, rgid, vid_valid, vid);
+	return ism_query_rgid(smcd->priv, rgid->gid, vid_valid, vid);
 }
 
 static int smcd_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb,

@@ -816,10 +816,11 @@ static int ism_signal_ieq(struct ism_dev *ism, u64 rgid, u32 trigger_irq,
 	return ism_cmd(ism, &cmd);
 }
 
-static int smcd_signal_ieq(struct smcd_dev *smcd, u64 rgid, u32 trigger_irq,
-			   u32 event_code, u64 info)
+static int smcd_signal_ieq(struct smcd_dev *smcd, struct smcd_gid *rgid,
+			   u32 trigger_irq, u32 event_code, u64 info)
 {
-	return ism_signal_ieq(smcd->priv, rgid, trigger_irq, event_code, info);
+	return ism_signal_ieq(smcd->priv, rgid->gid,
+			      trigger_irq, event_code, info);
 }
 
 static int smcd_move(struct smcd_dev *smcd, u64 dmb_tok, unsigned int idx,

@@ -840,9 +841,11 @@ static u64 ism_get_local_gid(struct ism_dev *ism)
 	return ism->local_gid;
 }
 
-static u64 smcd_get_local_gid(struct smcd_dev *smcd)
+static void smcd_get_local_gid(struct smcd_dev *smcd,
+			       struct smcd_gid *smcd_gid)
 {
-	return ism_get_local_gid(smcd->priv);
+	smcd_gid->gid = ism_get_local_gid(smcd->priv);
+	smcd_gid->gid_ext = 0;
 }
 
 static u16 ism_get_chid(struct ism_dev *ism)

include/net/smc.h +10 -5

index a002552be29c3..a0dc1187e96ed 100644
@@ -52,9 +52,14 @@ struct smcd_dmb {
 struct smcd_dev;
 struct ism_client;
 
+struct smcd_gid {
+	u64	gid;
+	u64	gid_ext;
+};
+
 struct smcd_ops {
-	int (*query_remote_gid)(struct smcd_dev *dev, u64 rgid, u32 vid_valid,
-				u32 vid);
+	int (*query_remote_gid)(struct smcd_dev *dev, struct smcd_gid *rgid,
+				u32 vid_valid, u32 vid);
 	int (*register_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb,
 			    struct ism_client *client);
 	int (*unregister_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb);

@@ -62,14 +67,14 @@ struct smcd_ops {
 	int (*del_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
 	int (*set_vlan_required)(struct smcd_dev *dev);
 	int (*reset_vlan_required)(struct smcd_dev *dev);
-	int (*signal_event)(struct smcd_dev *dev, u64 rgid, u32 trigger_irq,
-			    u32 event_code, u64 info);
+	int (*signal_event)(struct smcd_dev *dev, struct smcd_gid *rgid,
+			    u32 trigger_irq, u32 event_code, u64 info);
 	int (*move_data)(struct smcd_dev *dev, u64 dmb_tok, unsigned int idx,
 			 bool sf, unsigned int offset, void *data,
 			 unsigned int size);
 	int (*supports_v2)(void);
 	u8* (*get_system_eid)(void);
-	u64 (*get_local_gid)(struct smcd_dev *dev);
+	void (*get_local_gid)(struct smcd_dev *dev, struct smcd_gid *gid);
 	u16 (*get_chid)(struct smcd_dev *dev);
 	struct device* (*get_dev)(struct smcd_dev *dev);
 };

net/smc/af_smc.c +53 ~~-13~~

130 lines not on this page · raw message

net/smc/smc.h +0 -3

index e377980b84145..e0afef7a786f8 100644
@@ -29,9 +29,6 @@
 #define SMCPROTO_SMC		0	/* SMC protocol, IPv4 */
 #define SMCPROTO_SMC6		1	/* SMC protocol, IPv6 */
 
-#define SMC_MAX_ISM_DEVS	8	/* max # of proposed non-native ISM
-					 * devices
-					 */
 #define SMC_AUTOCORKING_DEFAULT_SIZE	0x10000	/* 64K by default */
 
 extern struct proto smc_proto;

net/smc/smc_clc.c +32 ~~-11~~

96 lines not on this page · raw message

net/smc/smc_clc.h +8 -2

index d7b1716cfb0fd..c8d6282ec9c04 100644
@@ -171,6 +171,11 @@ struct smc_clc_msg_proposal {	/* clc proposal message sent by Linux */
 
 #define SMC_CLC_MAX_V6_PREFIX		8
 #define SMC_CLC_MAX_UEID		8
+#define SMCD_CLC_MAX_V2_GID_ENTRIES	8 /* max # of CHID-GID entries in CLC
+					   * proposal SMC-Dv2 extension.
+					   * each ISM device takes one entry and
+					   * each virtual ISM takes two entries.
+					   */
 
 struct smc_clc_msg_proposal_area {
 	struct smc_clc_msg_proposal		pclc_base;

@@ -180,7 +185,8 @@ struct smc_clc_msg_proposal_area {
 	struct smc_clc_v2_extension		pclc_v2_ext;
 	u8			user_eids[SMC_CLC_MAX_UEID][SMC_MAX_EID_LEN];
 	struct smc_clc_smcd_v2_extension	pclc_smcd_v2_ext;
-	struct smc_clc_smcd_gid_chid		pclc_gidchids[SMC_MAX_ISM_DEVS];
+	struct smc_clc_smcd_gid_chid
+				pclc_gidchids[SMCD_CLC_MAX_V2_GID_ENTRIES];
 	struct smc_clc_msg_trail		pclc_trl;
 };

@@ -271,7 +277,7 @@ struct smc_clc_msg_accept_confirm {	/* clc accept / confirm message */
 			struct { /* v2 only, but 12 bytes reserved in v1 */
 				__be16 chid;
 				u8 eid[SMC_MAX_EID_LEN];
-				u8 reserved5[8];
+				__be64 gid_ext;
 			} __packed d1;
 		};
 	};

net/smc/smc_core.c +21 ~~-10~~

index f99bb9d0adcc6..605cdff671d65 100644
@@ -506,6 +506,7 @@ static int smc_nl_fill_smcd_lgr(struct smc_link_group *lgr,
 {
 	char smc_pnet[SMC_MAX_PNETID_LEN + 1];
 	struct smcd_dev *smcd = lgr->smcd;
+	struct smcd_gid smcd_gid;
 	struct nlattr *attrs;
 	void *nlh;

@@ -521,11 +522,11 @@ static int smc_nl_fill_smcd_lgr(struct smc_link_group *lgr,
 
 	if (nla_put_u32(skb, SMC_NLA_LGR_D_ID, *((u32 *)&lgr->id)))
 		goto errattr;
+	smcd->ops->get_local_gid(smcd, &smcd_gid);
 	if (nla_put_u64_64bit(skb, SMC_NLA_LGR_D_GID,
-			      smcd->ops->get_local_gid(smcd),
-				  SMC_NLA_LGR_D_PAD))
+			      smcd_gid.gid, SMC_NLA_LGR_D_PAD))
 		goto errattr;
-	if (nla_put_u64_64bit(skb, SMC_NLA_LGR_D_PEER_GID, lgr->peer_gid,
+	if (nla_put_u64_64bit(skb, SMC_NLA_LGR_D_PEER_GID, lgr->peer_gid.gid,
 			      SMC_NLA_LGR_D_PAD))
 		goto errattr;
 	if (nla_put_u8(skb, SMC_NLA_LGR_D_VLAN_ID, lgr->vlan_id))

@@ -876,7 +877,10 @@ static int smc_lgr_create(struct smc_sock *smc, struct smc_init_info *ini)
 		/* SMC-D specific settings */
 		smcd = ini->ism_dev[ini->ism_selected];
 		get_device(smcd->ops->get_dev(smcd));
-		lgr->peer_gid = ini->ism_peer_gid[ini->ism_selected];
+		lgr->peer_gid.gid =
+			ini->ism_peer_gid[ini->ism_selected].gid;
+		lgr->peer_gid.gid_ext =
+			ini->ism_peer_gid[ini->ism_selected].gid_ext;
 		lgr->smcd = ini->ism_dev[ini->ism_selected];
 		lgr_list = &ini->ism_dev[ini->ism_selected]->lgr_list;
 		lgr_lock = &lgr->smcd->lgr_lock;

@@ -1514,7 +1518,8 @@ void smc_lgr_terminate_sched(struct smc_link_group *lgr)
 }
 
 /* Called when peer lgr shutdown (regularly or abnormally) is received */
-void smc_smcd_terminate(struct smcd_dev *dev, u64 peer_gid, unsigned short vlan)
+void smc_smcd_terminate(struct smcd_dev *dev, struct smcd_gid *peer_gid,
+			unsigned short vlan)
 {
 	struct smc_link_group *lgr, *l;
 	LIST_HEAD(lgr_free_list);

@@ -1522,9 +1527,12 @@ void smc_smcd_terminate(struct smcd_dev *dev, u64 peer_gid, unsigned short vlan)
 	/* run common cleanup function and build free list */
 	spin_lock_bh(&dev->lgr_lock);
 	list_for_each_entry_safe(lgr, l, &dev->lgr_list, list) {
-		if ((!peer_gid || lgr->peer_gid == peer_gid) &&
+		if ((!peer_gid->gid ||
+		     (lgr->peer_gid.gid == peer_gid->gid &&
+		      !smc_ism_is_virtual(dev) ? 1 :
+		      lgr->peer_gid.gid_ext == peer_gid->gid_ext)) &&
 		    (vlan == VLAN_VID_MASK || lgr->vlan_id == vlan)) {
-			if (peer_gid) /* peer triggered termination */
+			if (peer_gid->gid) /* peer triggered termination */
 				lgr->peer_shutdown = 1;
 			list_move(&lgr->list, &lgr_free_list);
 			lgr->freeing = 1;

@@ -1860,9 +1868,12 @@ static bool smcr_lgr_match(struct smc_link_group *lgr, u8 smcr_version,
 }
 
 static bool smcd_lgr_match(struct smc_link_group *lgr,
-			   struct smcd_dev *smcismdev, u64 peer_gid)
+			   struct smcd_dev *smcismdev,
+			   struct smcd_gid *peer_gid)
 {
-	return lgr->peer_gid == peer_gid && lgr->smcd == smcismdev;
+	return lgr->peer_gid.gid == peer_gid->gid && lgr->smcd == smcismdev &&
+		smc_ism_is_virtual(smcismdev) ?
+		(lgr->peer_gid.gid_ext == peer_gid->gid_ext) : 1;
 }
 
 /* create a new SMC connection (and a new link group if necessary) */

@@ -1892,7 +1903,7 @@ int smc_conn_create(struct smc_sock *smc, struct smc_init_info *ini)
 		write_lock_bh(&lgr->conns_lock);
 		if ((ini->is_smcd ?
 		     smcd_lgr_match(lgr, ini->ism_dev[ini->ism_selected],
-				    ini->ism_peer_gid[ini->ism_selected]) :
+				    &ini->ism_peer_gid[ini->ism_selected]) :
 		     smcr_lgr_match(lgr, ini->smcr_version,
 				    ini->peer_systemid,
 				    ini->peer_gid, ini->peer_mac, role,

net/smc/smc_core.h +12 -5

index 120027d404692..670f8359da558 100644
@@ -17,9 +17,11 @@
 #include <linux/pci.h>
 #include <rdma/ib_verbs.h>
 #include <net/genetlink.h>
+#include <net/smc.h>
 
 #include "smc.h"
 #include "smc_ib.h"
+#include "smc_clc.h"
 
 #define SMC_RMBS_PER_LGR_MAX	255	/* max. # of RMBs per link group */
 #define SMC_CONN_PER_LGR_MIN	16	/* min. # of connections per link group */

@@ -355,7 +357,7 @@ struct smc_link_group {
 						/* max links can be added in lgr */
 		};
 		struct { /* SMC-D */
-			u64			peer_gid;
+			struct smcd_gid		peer_gid;
 						/* Peer GID (remote) */
 			struct smcd_dev		*smcd;
 						/* ISM device for VLAN reg. */

@@ -392,6 +394,11 @@ struct smc_init_info_smcrv2 {
 	struct smc_gidlist	gidlist;
 };
 
+#define SMC_MAX_V2_ISM_DEVS	SMCD_CLC_MAX_V2_GID_ENTRIES
+				/* max # of proposed non-native ISM devices,
+				 * which can't exceed the max # of CHID-GID
+				 * entries in CLC proposal SMC-Dv2 extension.
+				 */
 struct smc_init_info {
 	u8			is_smcd;
 	u8			smc_type_v1;

@@ -416,9 +423,9 @@ struct smc_init_info {
 	u32			ib_clcqpn;
 	struct smc_init_info_smcrv2 smcrv2;
 	/* SMC-D */
-	u64			ism_peer_gid[SMC_MAX_ISM_DEVS + 1];
-	struct smcd_dev		*ism_dev[SMC_MAX_ISM_DEVS + 1];
-	u16			ism_chid[SMC_MAX_ISM_DEVS + 1];
+	struct smcd_gid		ism_peer_gid[SMC_MAX_V2_ISM_DEVS + 1];
+	struct smcd_dev		*ism_dev[SMC_MAX_V2_ISM_DEVS + 1];
+	u16			ism_chid[SMC_MAX_V2_ISM_DEVS + 1];
 	u8			ism_offered_cnt; /* # of ISM devices offered */
 	u8			ism_selected;    /* index of selected ISM dev*/
 	u8			smcd_version;

@@ -544,7 +551,7 @@ void smc_lgr_hold(struct smc_link_group *lgr);
 void smc_lgr_put(struct smc_link_group *lgr);
 void smcr_port_add(struct smc_ib_device *smcibdev, u8 ibport);
 void smcr_port_err(struct smc_ib_device *smcibdev, u8 ibport);
-void smc_smcd_terminate(struct smcd_dev *dev, u64 peer_gid,
+void smc_smcd_terminate(struct smcd_dev *dev, struct smcd_gid *peer_gid,
 			unsigned short vlan);
 void smc_smcd_terminate_all(struct smcd_dev *dev);
 void smc_smcr_terminate_all(struct smc_ib_device *smcibdev);

net/smc/smc_diag.c +5 -2

index d58c699b5328a..086e48b3cdea6 100644
@@ -21,6 +21,7 @@
 
 #include "smc.h"
 #include "smc_core.h"
+#include "smc_ism.h"
 
 struct smc_diag_dump_ctx {
 	int pos[2];

@@ -167,12 +168,14 @@ static int __smc_diag_dump(struct sock *sk, struct sk_buff *skb,
 		struct smc_connection *conn = &smc->conn;
 		struct smcd_diag_dmbinfo dinfo;
 		struct smcd_dev *smcd = conn->lgr->smcd;
+		struct smcd_gid smcd_gid;
 
 		memset(&dinfo, 0, sizeof(dinfo));
 
 		dinfo.linkid = *((u32 *)conn->lgr->id);
-		dinfo.peer_gid = conn->lgr->peer_gid;
-		dinfo.my_gid = smcd->ops->get_local_gid(smcd);
+		dinfo.peer_gid = conn->lgr->peer_gid.gid;
+		smcd->ops->get_local_gid(smcd, &smcd_gid);
+		dinfo.my_gid = smcd_gid.gid;
 		dinfo.token = conn->rmb_desc->token;
 		dinfo.peer_token = conn->peer_token;

net/smc/smc_ism.c +11 -6

index fbee2493091f1..a33f861cf7c19 100644
@@ -44,7 +44,8 @@ static struct ism_client smc_ism_client = {
 #endif
 
 /* Test if an ISM communication is possible - same CPC */
-int smc_ism_cantalk(u64 peer_gid, unsigned short vlan_id, struct smcd_dev *smcd)
+int smc_ism_cantalk(struct smcd_gid *peer_gid, unsigned short vlan_id,
+		    struct smcd_dev *smcd)
 {
 	return smcd->ops->query_remote_gid(smcd, peer_gid, vlan_id ? 1 : 0,
 					   vlan_id);

@@ -208,7 +209,7 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 	dmb.dmb_len = dmb_len;
 	dmb.sba_idx = dmb_desc->sba_idx;
 	dmb.vlan_id = lgr->vlan_id;
-	dmb.rgid = lgr->peer_gid;
+	dmb.rgid = lgr->peer_gid.gid;
 	rc = lgr->smcd->ops->register_dmb(lgr->smcd, &dmb, &smc_ism_client);
 	if (!rc) {
 		dmb_desc->sba_idx = dmb.sba_idx;

@@ -340,18 +341,20 @@ union smcd_sw_event_info {
 
 static void smcd_handle_sw_event(struct smc_ism_event_work *wrk)
 {
+	struct smcd_gid peer_gid = { .gid = wrk->event.tok,
+				     .gid_ext = 0 };
 	union smcd_sw_event_info ev_info;
 
 	ev_info.info = wrk->event.info;
 	switch (wrk->event.code) {
 	case ISM_EVENT_CODE_SHUTDOWN:	/* Peer shut down DMBs */
-		smc_smcd_terminate(wrk->smcd, wrk->event.tok, ev_info.vlan_id);
+		smc_smcd_terminate(wrk->smcd, &peer_gid, ev_info.vlan_id);
 		break;
 	case ISM_EVENT_CODE_TESTLINK:	/* Activity timer */
 		if (ev_info.code == ISM_EVENT_REQUEST) {
 			ev_info.code = ISM_EVENT_RESPONSE;
 			wrk->smcd->ops->signal_event(wrk->smcd,
-						     wrk->event.tok,
+						     &peer_gid,
 						     ISM_EVENT_REQUEST_IR,
 						     ISM_EVENT_CODE_TESTLINK,
 						     ev_info.info);

@@ -365,10 +368,12 @@ static void smc_ism_event_work(struct work_struct *work)
 {
 	struct smc_ism_event_work *wrk =
 		container_of(work, struct smc_ism_event_work, work);
+	struct smcd_gid smcd_gid = { .gid = wrk->event.tok,
+				     .gid_ext = 0 };
 
 	switch (wrk->event.type) {
 	case ISM_EVENT_GID:	/* GID event, token is peer GID */
-		smc_smcd_terminate(wrk->smcd, wrk->event.tok, VLAN_VID_MASK);
+		smc_smcd_terminate(wrk->smcd, &smcd_gid, VLAN_VID_MASK);
 		break;
 	case ISM_EVENT_DMB:
 		break;

@@ -525,7 +530,7 @@ int smc_ism_signal_shutdown(struct smc_link_group *lgr)
 	memcpy(ev_info.uid, lgr->id, SMC_LGR_ID_SIZE);
 	ev_info.vlan_id = lgr->vlan_id;
 	ev_info.code = ISM_EVENT_REQUEST;
-	rc = lgr->smcd->ops->signal_event(lgr->smcd, lgr->peer_gid,
+	rc = lgr->smcd->ops->signal_event(lgr->smcd, &lgr->peer_gid,
 					  ISM_EVENT_REQUEST_IR,
 					  ISM_EVENT_CODE_SHUTDOWN,
 					  ev_info.info);

net/smc/smc_ism.h +2 -1

index d1228a615f23c..0e5e563099ec3 100644
@@ -32,7 +32,8 @@ struct smc_ism_vlanid {			/* VLAN id set on ISM device */
 
 struct smcd_dev;
 
-int smc_ism_cantalk(u64 peer_gid, unsigned short vlan_id, struct smcd_dev *dev);
+int smc_ism_cantalk(struct smcd_gid *peer_gid, unsigned short vlan_id,
+		    struct smcd_dev *dev);
 void smc_ism_set_conn(struct smc_connection *conn);
 void smc_ism_unset_conn(struct smc_connection *conn);
 int smc_ism_get_vlan(struct smcd_dev *dev, unsigned short vlan_id);

net/smc/smc_pnet.c +2 -2

index 284cec1e20ec1..dbcc72b43d0c0 100644
@@ -1113,8 +1113,8 @@ static void smc_pnet_find_ism_by_pnetid(struct net_device *ndev,
 	list_for_each_entry(ismdev, &smcd_dev_list.list, list) {
 		if (smc_pnet_match(ismdev->pnetid, ndev_pnetid) &&
 		    !ismdev->going_away &&
-		    (!ini->ism_peer_gid[0] ||
-		     !smc_ism_cantalk(ini->ism_peer_gid[0], ini->vlan_id,
+		    (!ini->ism_peer_gid[0].gid ||
+		     !smc_ism_cantalk(&ini->ism_peer_gid[0], ini->vlan_id,
 				      ismdev))) {
 			ini->ism_dev[0] = ismdev;
 			break;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #39 lore raw

PATCH 39/356 net/smc: mark optional smcd_ops and check for support when called

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit d1d8d0b6c7c68b0665456831fa779174ebd78f90 ]

Some operations are not supported by new introduced Emulated-ISM, so mark them as optional and check if the device supports them when called.

Signed-off-by Wen Gu <[email protected]>

Reviewed-and-tested-by Jan Karcher <[email protected]>

Signed-off-by Paolo Abeni <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/smc.h | 14 ++++++++------
 net/smc/smc_ism.c |  9 ++++++++-
 2 files changed, 16 insertions(+), 7 deletions(-)

include/net/smc.h +8 -6

index a0dc1187e96ed..9dfe57f3e4f0b 100644
@@ -63,12 +63,6 @@ struct smcd_ops {
 	int (*register_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb,
 			    struct ism_client *client);
 	int (*unregister_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb);
-	int (*add_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
-	int (*del_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
-	int (*set_vlan_required)(struct smcd_dev *dev);
-	int (*reset_vlan_required)(struct smcd_dev *dev);
-	int (*signal_event)(struct smcd_dev *dev, struct smcd_gid *rgid,
-			    u32 trigger_irq, u32 event_code, u64 info);
 	int (*move_data)(struct smcd_dev *dev, u64 dmb_tok, unsigned int idx,
 			 bool sf, unsigned int offset, void *data,
 			 unsigned int size);

@@ -77,6 +71,14 @@ struct smcd_ops {
 	void (*get_local_gid)(struct smcd_dev *dev, struct smcd_gid *gid);
 	u16 (*get_chid)(struct smcd_dev *dev);
 	struct device* (*get_dev)(struct smcd_dev *dev);
+
+	/* optional operations */
+	int (*add_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
+	int (*del_vlan_id)(struct smcd_dev *dev, u64 vlan_id);
+	int (*set_vlan_required)(struct smcd_dev *dev);
+	int (*reset_vlan_required)(struct smcd_dev *dev);
+	int (*signal_event)(struct smcd_dev *dev, struct smcd_gid *rgid,
+			    u32 trigger_irq, u32 event_code, u64 info);
 };
 
 struct smcd_dev {

net/smc/smc_ism.c +8 -1

index a33f861cf7c19..3623df320de55 100644
@@ -105,6 +105,8 @@ int smc_ism_get_vlan(struct smcd_dev *smcd, unsigned short vlanid)
 
 	if (!vlanid)			/* No valid vlan id */
 		return -EINVAL;
+	if (!smcd->ops->add_vlan_id)
+		return -EOPNOTSUPP;
 
 	/* create new vlan entry, in case we need it */
 	new_vlan = kzalloc(sizeof(*new_vlan), GFP_KERNEL);

@@ -150,6 +152,8 @@ int smc_ism_put_vlan(struct smcd_dev *smcd, unsigned short vlanid)
 
 	if (!vlanid)			/* No valid vlan id */
 		return -EINVAL;
+	if (!smcd->ops->del_vlan_id)
+		return -EOPNOTSUPP;
 
 	spin_lock_irqsave(&smcd->lock, flags);
 	list_for_each_entry(vlan, &smcd->vlan, list) {

@@ -351,7 +355,8 @@ static void smcd_handle_sw_event(struct smc_ism_event_work *wrk)
 		smc_smcd_terminate(wrk->smcd, &peer_gid, ev_info.vlan_id);
 		break;
 	case ISM_EVENT_CODE_TESTLINK:	/* Activity timer */
-		if (ev_info.code == ISM_EVENT_REQUEST) {
+		if (ev_info.code == ISM_EVENT_REQUEST &&
+		    wrk->smcd->ops->signal_event) {
 			ev_info.code = ISM_EVENT_RESPONSE;
 			wrk->smcd->ops->signal_event(wrk->smcd,
 						     &peer_gid,

@@ -526,6 +531,8 @@ int smc_ism_signal_shutdown(struct smc_link_group *lgr)
 
 	if (lgr->peer_shutdown)
 		return 0;
+	if (!lgr->smcd->ops->signal_event)
+		return 0;
 
 	memcpy(ev_info.uid, lgr->id, SMC_LGR_ID_SIZE);
 	ev_info.vlan_id = lgr->vlan_id;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #40 lore raw

PATCH 40/356 net/smc: add operations to merge sndbuf with peer DMB

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit 4398888268582cb51b69c6ee94f551bb8d37d12f ]

In some scenarios using Emulated-ISM device, sndbuf can share the same physical memory region with peer DMB to avoid data copy from one side to the other. In such case the sndbuf is only a descriptor that describes the shared memory and does not actually occupy memory, it's more like a ghost buffer.

      +----------+                     +----------+
      | socket A |                     | socket B |
      +----------+                     +----------+
            |                               |
       +--------+                       +--------+
       | sndbuf |                       |  DMB   |
       |  desc  |                       |  desc  |
       +--------+                       +--------+
            |                               |
            |                          +----v-----+
            +-------------------------->  memory  |
                                       +----------+

So here introduces three new SMC-D device operations to check if this feature is supported by device, and to {attach|detach} ghost sndbuf to peer DMB. For now only loopback-ism supports this.

Signed-off-by Wen Gu <[email protected]>

Reviewed-and-tested-by Jan Karcher <[email protected]>

Signed-off-by Paolo Abeni <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/smc.h |  3 +++
 net/smc/smc_ism.c | 40 ++++++++++++++++++++++++++++++++++++++++
 net/smc/smc_ism.h |  4 ++++
 3 files changed, 47 insertions(+)

include/net/smc.h +3 -0

index 9dfe57f3e4f0b..6fef76087b9ed 100644
@@ -79,6 +79,9 @@ struct smcd_ops {
 	int (*reset_vlan_required)(struct smcd_dev *dev);
 	int (*signal_event)(struct smcd_dev *dev, struct smcd_gid *rgid,
 			    u32 trigger_irq, u32 event_code, u64 info);
+	int (*support_dmb_nocopy)(struct smcd_dev *dev);
+	int (*attach_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb);
+	int (*detach_dmb)(struct smcd_dev *dev, u64 token);
 };
 
 struct smcd_dev {

net/smc/smc_ism.c +40 -0

index 3623df320de55..61ffc72014013 100644
@@ -228,6 +228,46 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len,
 #endif
 }
 
+bool smc_ism_support_dmb_nocopy(struct smcd_dev *smcd)
+{
+	/* for now only loopback-ism supports
+	 * merging sndbuf with peer DMB to avoid
+	 * data copies between them.
+	 */
+	return (smcd->ops->support_dmb_nocopy &&
+		smcd->ops->support_dmb_nocopy(smcd));
+}
+
+int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token,
+		       struct smc_buf_desc *dmb_desc)
+{
+	struct smcd_dmb dmb;
+	int rc = 0;
+
+	if (!dev->ops->attach_dmb)
+		return -EINVAL;
+
+	memset(&dmb, 0, sizeof(dmb));
+	dmb.dmb_tok = token;
+	rc = dev->ops->attach_dmb(dev, &dmb);
+	if (!rc) {
+		dmb_desc->sba_idx = dmb.sba_idx;
+		dmb_desc->token = dmb.dmb_tok;
+		dmb_desc->cpu_addr = dmb.cpu_addr;
+		dmb_desc->dma_addr = dmb.dma_addr;
+		dmb_desc->len = dmb.dmb_len;
+	}
+	return rc;
+}
+
+int smc_ism_detach_dmb(struct smcd_dev *dev, u64 token)
+{
+	if (!dev->ops->detach_dmb)
+		return -EINVAL;
+
+	return dev->ops->detach_dmb(dev, token);
+}
+
 static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd,
 				  struct sk_buff *skb,
 				  struct netlink_callback *cb)

net/smc/smc_ism.h +4 -0

index 0e5e563099ec3..8312c3586d2b3 100644
@@ -41,6 +41,10 @@ int smc_ism_put_vlan(struct smcd_dev *dev, unsigned short vlan_id);
 int smc_ism_register_dmb(struct smc_link_group *lgr, int buf_size,
 			 struct smc_buf_desc *dmb_desc);
 int smc_ism_unregister_dmb(struct smcd_dev *dev, struct smc_buf_desc *dmb_desc);
+bool smc_ism_support_dmb_nocopy(struct smcd_dev *smcd);
+int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token,
+		       struct smc_buf_desc *dmb_desc);
+int smc_ism_detach_dmb(struct smcd_dev *dev, u64 token);
 int smc_ism_signal_shutdown(struct smc_link_group *lgr);
 void smc_ism_get_system_eid(u8 **eid);
 u16 smc_ism_get_chid(struct smcd_dev *dev);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #41 lore raw

PATCH 41/356 net/smc: {at|de}tach sndbuf to peer DMB if supported

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit ae2be35cbed2c8385e890147ea321a3fcc3ca5fa ]

If the device used by SMC-D supports merging local sndbuf to peer DMB, then create sndbuf descriptor and attach it to peer DMB once peer token is obtained, and detach and free the sndbuf descriptor when the connection is freed.

Signed-off-by Wen Gu <[email protected]>

Reviewed-and-tested-by Jan Karcher <[email protected]>

Signed-off-by Paolo Abeni <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/af_smc.c   | 16 ++++++++++++
 net/smc/smc_core.c | 61 +++++++++++++++++++++++++++++++++++++++++++++-
 net/smc/smc_core.h |  1 +
 3 files changed, 77 insertions(+), 1 deletion(-)

net/smc/af_smc.c +16 -0

index f3ed53ae849d3..c4b30ea4b6ca0 100644
@@ -1418,6 +1418,14 @@ static int smc_connect_ism(struct smc_sock *smc,
 	}
 
 	smc_conn_save_peer_info(smc, aclc);
+
+	if (smc_ism_support_dmb_nocopy(smc->conn.lgr->smcd)) {
+		rc = smcd_buf_attach(smc);
+		if (rc) {
+			rc = SMC_CLC_DECL_MEM;	/* try to fallback */
+			goto connect_abort;
+		}
+	}
 	smc_close_init(smc);
 	smc_rx_init(smc);
 	smc_tx_init(smc);

@@ -2522,6 +2530,14 @@ static void smc_listen_work(struct work_struct *work)
 		mutex_unlock(&smc_server_lgr_pending);
 	}
 	smc_conn_save_peer_info(new_smc, cclc);
+
+	if (ini->is_smcd &&
+	    smc_ism_support_dmb_nocopy(new_smc->conn.lgr->smcd)) {
+		rc = smcd_buf_attach(new_smc);
+		if (rc)
+			goto out_decl;
+	}
+
 	smc_listen_out_connected(new_smc);
 	SMC_STAT_SERV_SUCC_INC(sock_net(newclcsock->sk), ini);
 	goto out_free;

net/smc/smc_core.c +60 -1

96 lines not on this page · raw message

net/smc/smc_core.h +1 -0

index 670f8359da558..de001f4b46c7d 100644
@@ -556,6 +556,7 @@ void smc_smcd_terminate(struct smcd_dev *dev, struct smcd_gid *peer_gid,
 void smc_smcd_terminate_all(struct smcd_dev *dev);
 void smc_smcr_terminate_all(struct smc_ib_device *smcibdev);
 int smc_buf_create(struct smc_sock *smc, bool is_smcd);
+int smcd_buf_attach(struct smc_sock *smc);
 int smc_uncompress_bufsize(u8 compressed);
 int smc_rmb_rtoken_handling(struct smc_connection *conn, struct smc_link *link,
 			    struct smc_clc_msg_accept_confirm *clc);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #42 lore raw

PATCH 42/356 net/smc: refactoring initialization of smc sock

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: D. Wythe <[email protected]>

[ Upstream commit d0e35656d83458d668593930f1568d464dde429c ]

This patch aims to isolate the shared components of SMC socket allocation by introducing smc_sk_init() for sock initialization and __smc_create_clcsk() for the initialization of clcsock.

This is in preparation for the subsequent implementation of the AF_INET version of SMC.

Signed-off-by D. Wythe <[email protected]>

Reviewed-by Tony Lu <[email protected]>

Reviewed-by Dust Li <[email protected]>

Tested-by Niklas Schnelle <[email protected]>

Tested-by Wenjia Zhang <[email protected]>

Stable-dep-of: 0541db8ee32c ("net/smc: initialize close_work early to avoid warning")

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/af_smc.c | 86 +++++++++++++++++++++++++++---------------------
 net/smc/smc.h    |  5 +++
 2 files changed, 53 insertions(+), 38 deletions(-)

net/smc/af_smc.c +48 ~~-38~~

index c4b30ea4b6ca0..f343e91eec0e4 100644
@@ -362,25 +362,15 @@ static void smc_destruct(struct sock *sk)
 		return;
 }
 
-static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,
-				   int protocol)
+void smc_sk_init(struct net *net, struct sock *sk, int protocol)
 {
-	struct smc_sock *smc;
-	struct proto *prot;
-	struct sock *sk;
-
-	prot = (protocol == SMCPROTO_SMC6) ? &smc_proto6 : &smc_proto;
-	sk = sk_alloc(net, PF_SMC, GFP_KERNEL, prot, 0);
-	if (!sk)
-		return NULL;
+	struct smc_sock *smc = smc_sk(sk);
 
-	sock_init_data(sock, sk); /* sets sk_refcnt to 1 */
 	sk->sk_state = SMC_INIT;
 	sk->sk_destruct = smc_destruct;
 	sk->sk_protocol = protocol;
 	WRITE_ONCE(sk->sk_sndbuf, 2 * READ_ONCE(net->smc.sysctl_wmem));
 	WRITE_ONCE(sk->sk_rcvbuf, 2 * READ_ONCE(net->smc.sysctl_rmem));
-	smc = smc_sk(sk);
 	INIT_WORK(&smc->tcp_listen_work, smc_tcp_listen_work);
 	INIT_WORK(&smc->connect_work, smc_connect_work);
 	INIT_DELAYED_WORK(&smc->conn.tx_work, smc_tx_work);

@@ -390,6 +380,24 @@ static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,
 	sk->sk_prot->hash(sk);
 	mutex_init(&smc->clcsock_release_lock);
 	smc_init_saved_callbacks(smc);
+	smc->limit_smc_hs = net->smc.limit_smc_hs;
+	smc->use_fallback = false; /* assume rdma capability first */
+	smc->fallback_rsn = 0;
+}
+
+static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,
+				   int protocol)
+{
+	struct proto *prot;
+	struct sock *sk;
+
+	prot = (protocol == SMCPROTO_SMC6) ? &smc_proto6 : &smc_proto;
+	sk = sk_alloc(net, PF_SMC, GFP_KERNEL, prot, 0);
+	if (!sk)
+		return NULL;
+
+	sock_init_data(sock, sk); /* sets sk_refcnt to 1 */
+	smc_sk_init(net, sk, protocol);
 
 	return sk;
 }

@@ -3303,6 +3311,31 @@ static const struct proto_ops smc_sock_ops = {
 	.splice_read	= smc_splice_read,
 };
 
+int smc_create_clcsk(struct net *net, struct sock *sk, int family)
+{
+	struct smc_sock *smc = smc_sk(sk);
+	int rc;
+
+	rc = sock_create_kern(net, family, SOCK_STREAM, IPPROTO_TCP,
+			      &smc->clcsock);
+	if (rc) {
+		sk_common_release(sk);
+		return rc;
+	}
+
+	/* smc_clcsock_release() does not wait smc->clcsock->sk's
+	 * destruction;  its sk_state might not be TCP_CLOSE after
+	 * smc->sk is close()d, and TCP timers can be fired later,
+	 * which need net ref.
+	 */
+	sk = smc->clcsock->sk;
+	__netns_tracker_free(net, &sk->ns_tracker, false);
+	sk->sk_net_refcnt = 1;
+	get_net_track(net, &sk->ns_tracker, GFP_KERNEL);
+	sock_inuse_add(net, 1);
+	return 0;
+}
+
 static int __smc_create(struct net *net, struct socket *sock, int protocol,
 			int kern, struct socket *clcsock)
 {

@@ -3328,35 +3361,12 @@ static int __smc_create(struct net *net, struct socket *sock, int protocol,
 
 	/* create internal TCP socket for CLC handshake and fallback */
 	smc = smc_sk(sk);
-	smc->use_fallback = false; /* assume rdma capability first */
-	smc->fallback_rsn = 0;
-
-	/* default behavior from limit_smc_hs in every net namespace */
-	smc->limit_smc_hs = net->smc.limit_smc_hs;
 
 	rc = 0;
-	if (!clcsock) {
-		rc = sock_create_kern(net, family, SOCK_STREAM, IPPROTO_TCP,
-				      &smc->clcsock);
-		if (rc) {
-			sk_common_release(sk);
-			goto out;
-		}
-
-		/* smc_clcsock_release() does not wait smc->clcsock->sk's
-		 * destruction;  its sk_state might not be TCP_CLOSE after
-		 * smc->sk is close()d, and TCP timers can be fired later,
-		 * which need net ref.
-		 */
-		sk = smc->clcsock->sk;
-		__netns_tracker_free(net, &sk->ns_tracker, false);
-		sk->sk_net_refcnt = 1;
-		get_net_track(net, &sk->ns_tracker, GFP_KERNEL);
-		sock_inuse_add(net, 1);
-	} else {
+	if (clcsock)
 		smc->clcsock = clcsock;
-	}
-
+	else
+		rc = smc_create_clcsk(net, sk, family);
 out:
 	return rc;
 }

net/smc/smc.h +5 -0

index e0afef7a786f8..36699ba551887 100644
@@ -34,6 +34,11 @@
 extern struct proto smc_proto;
 extern struct proto smc_proto6;
 
+/* smc sock initialization */
+void smc_sk_init(struct net *net, struct sock *sk, int protocol);
+/* clcsock initialization */
+int smc_create_clcsk(struct net *net, struct sock *sk, int family);
+
 #ifdef ATOMIC64_INIT
 #define KERNEL_HAS_ATOMIC64
 #endif

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #43 lore raw

PATCH 43/356 net/smc: initialize close_work early to avoid warning

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit 0541db8ee32c09463a72d0987382b3a3336b0043 ]

We encountered a warning that close_work was canceled before initialization.

  WARNING: CPU: 7 PID: 111103 at kernel/workqueue.c:3047 __flush_work+0x19e/0x1b0
  Workqueue: events smc_lgr_terminate_work [smc]
  RIP: 0010:__flush_work+0x19e/0x1b0
  Call Trace:
   ? __wake_up_common+0x7a/0x190
   ? work_busy+0x80/0x80
   __cancel_work_timer+0xe3/0x160
   smc_close_cancel_work+0x1a/0x70 [smc]
   smc_close_active_abort+0x207/0x360 [smc]
   __smc_lgr_terminate.part.38+0xc8/0x180 [smc]
   process_one_work+0x19e/0x340
   worker_thread+0x30/0x370
   ? process_one_work+0x340/0x340
   kthread+0x117/0x130
   ? __kthread_cancel_work+0x50/0x50
   ret_from_fork+0x22/0x30

This is because when smc_close_cancel_work is triggered, e.g. the RDMA driver is rmmod and the LGR is terminated, the conn->close_work is flushed before initialization, resulting in WARN_ON(!work->func).

__smc_lgr_terminate | smc_connect_{rdma|ism}
-------------------------------------------------------------
| smc_conn_create | \- smc_lgr_register_conn

for conn in lgr->conns_all      |
\- smc_conn_kill                |
   \- smc_close_active_abort    |
      \- smc_close_cancel_work  |
         \- cancel_work_sync    |
            \- __flush_work     |
	         (close_work)   |
	                        | smc_close_init
	                        | \- INIT_WORK(&close_work)

So fix this by initializing close_work before establishing the connection.

Fixes 46c28dbd4c23 ("net/smc: no socket state changes in tasklet context")

Fixes 413498440e30 ("net/smc: add SMC-D support in af_smc")

Signed-off-by Wen Gu <[email protected]>

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/af_smc.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

net/smc/af_smc.c +1 -3

index f343e91eec0e4..755659703a625 100644
@@ -383,6 +383,7 @@ void smc_sk_init(struct net *net, struct sock *sk, int protocol)
 	smc->limit_smc_hs = net->smc.limit_smc_hs;
 	smc->use_fallback = false; /* assume rdma capability first */
 	smc->fallback_rsn = 0;
+	smc_close_init(smc);
 }
 
 static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,

@@ -1298,7 +1299,6 @@ static int smc_connect_rdma(struct smc_sock *smc,
 		goto connect_abort;
 	}
 
-	smc_close_init(smc);
 	smc_rx_init(smc);
 
 	if (ini->first_contact_local) {

@@ -1434,7 +1434,6 @@ static int smc_connect_ism(struct smc_sock *smc,
 			goto connect_abort;
 		}
 	}
-	smc_close_init(smc);
 	smc_rx_init(smc);
 	smc_tx_init(smc);

@@ -2486,7 +2485,6 @@ static void smc_listen_work(struct work_struct *work)
 		goto out_decl;
 
 	mutex_lock(&smc_server_lgr_pending);
-	smc_close_init(new_smc);
 	smc_rx_init(new_smc);
 	smc_tx_init(new_smc);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #44 lore raw

PATCH 44/356 net/smc: fix LGR and link use-after-free issue

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit 2c7f14ed9c19ec0f149479d1c2842ec1f9bf76d7 ]

We encountered a LGR/link use-after-free issue, which manifested as the LGR/link refcnt reaching 0 early and entering the clear process, making resource access unsafe.

refcount_t: addition on 0; use-after-free.

 WARNING: CPU: 14 PID: 107447 at lib/refcount.c:25 refcount_warn_saturate+0x9c/0x140
 Workqueue: events smc_lgr_terminate_work [smc]
 Call trace:
  refcount_warn_saturate+0x9c/0x140
  __smc_lgr_terminate.part.45+0x2a8/0x370 [smc]
  smc_lgr_terminate_work+0x28/0x30 [smc]
  process_one_work+0x1b8/0x420
  worker_thread+0x158/0x510
  kthread+0x114/0x118

refcount_t: underflow; use-after-free.

 WARNING: CPU: 6 PID: 93140 at lib/refcount.c:28 refcount_warn_saturate+0xf0/0x140
 Workqueue: smc_hs_wq smc_listen_work [smc]
 Call trace:
  refcount_warn_saturate+0xf0/0x140
  smcr_link_put+0x1cc/0x1d8 [smc]
  smc_conn_free+0x110/0x1b0 [smc]
  smc_conn_abort+0x50/0x60 [smc]
  smc_listen_find_device+0x75c/0x790 [smc]
  smc_listen_work+0x368/0x8a0 [smc]
  process_one_work+0x1b8/0x420
  worker_thread+0x158/0x510
  kthread+0x114/0x118

It is caused by repeated release of LGR/link refcnt. One suspect is that smc_conn_free() is called repeatedly because some smc_conn_free() from server listening path are not protected by sock lock.

e.g.

Calls under socklock | smc_listen_work
-------------------------------------------------------

lock_sock(sk)               | smc_conn_abort
smc_conn_free               | \- smc_conn_free
\- smcr_link_put            |    \- smcr_link_put (duplicated)

release_sock(sk)

So here add sock lock protection in smc_listen_work() path, making it exclusive with other connection operations.

Fixes 3b2dec2603d5 ("net/smc: restructure client and server code in af_smc")

Co-developed-by Guangguan Wang <[email protected]>

Signed-off-by Guangguan Wang <[email protected]>

Co-developed-by Kai <[email protected]>

Signed-off-by Kai <[email protected]>

Signed-off-by Wen Gu <[email protected]>

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/smc/af_smc.c | 2 ++
 1 file changed, 2 insertions(+)

net/smc/af_smc.c +2 -0

index 755659703a625..77c6c0dff069e 100644
@@ -1907,6 +1907,7 @@ static void smc_listen_out(struct smc_sock *new_smc)
 	if (tcp_sk(new_smc->clcsock->sk)->syn_smc)
 		atomic_dec(&lsmc->queued_smc_hs);
 
+	release_sock(newsmcsk); /* lock in smc_listen_work() */
 	if (lsmc->sk.sk_state == SMC_LISTEN) {
 		lock_sock_nested(&lsmc->sk, SINGLE_DEPTH_NESTING);
 		smc_accept_enqueue(&lsmc->sk, newsmcsk);

@@ -2428,6 +2429,7 @@ static void smc_listen_work(struct work_struct *work)
 	u8 accept_version;
 	int rc = 0;
 
+	lock_sock(&new_smc->sk); /* release in smc_listen_out() */
 	if (new_smc->listen_smc->sk.sk_state != SMC_LISTEN)
 		return smc_listen_out_err(new_smc);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #45 lore raw

PATCH 45/356 net/qed: allow old cards not supporting "num_images" to work

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Louis Leseur <[email protected]>

[ Upstream commit 7a0ea70da56ee8c2716d0b79e9959d3c47efab62 ]

Commit 43645ce03e00 ("qed: Populate nvm image attribute shadow.") added support for populating flash image attributes, notably "num_images". However, some cards were not able to return this information. In such cases, the driver would return EINVAL, causing the driver to exit.

Add check to return EOPNOTSUPP instead of EINVAL when the card is not able to return these information. The caller function already handles EOPNOTSUPP without error.

Fixes 43645ce03e00 ("qed: Populate nvm image attribute shadow.")

Co-developed-by Florian Forestier <[email protected]>

Signed-off-by Florian Forestier <[email protected]>

Signed-off-by Louis Leseur <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/qlogic/qed/qed_mcp.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

drivers/net/ethernet/qlogic/qed/qed_mcp.c +3 -1

index 16e6bd4661433..6218d9c268554 100644
@@ -3314,7 +3314,9 @@ int qed_mcp_bist_nvm_get_num_images(struct qed_hwfn *p_hwfn,
 	if (rc)
 		return rc;
 
-	if (((rsp & FW_MSG_CODE_MASK) != FW_MSG_CODE_OK))
+	if (((rsp & FW_MSG_CODE_MASK) == FW_MSG_CODE_UNSUPPORTED))
+		rc = -EOPNOTSUPP;
+	else if (((rsp & FW_MSG_CODE_MASK) != FW_MSG_CODE_OK))
 		rc = -EINVAL;
 
 	return rc;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #46 lore raw

PATCH 46/356 ixgbevf: stop attempting IPSEC offload on Mailbox API 1.5

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jacob Keller <[email protected]>

[ Upstream commit d0725312adf5a803de8f621bd1b12ba7a6464a29 ]

Commit 339f28964147 ("ixgbevf: Add support for new mailbox communication between PF and VF") added support for v1.5 of the PF to VF mailbox communication API. This commit mistakenly enabled IPSEC offload for API v1.5.

No implementation of the v1.5 API has support for IPSEC offload. This offload is only supported by the Linux PF as mailbox API v1.4. In fact, the v1.5 API is not implemented in any Linux PF.

Attempting to enable IPSEC offload on a PF which supports v1.5 API will not work. Only the Linux upstream ixgbe and ixgbevf support IPSEC offload, and only as part of the v1.4 API.

Fix the ixgbevf Linux driver to stop attempting IPSEC offload when the mailbox API does not support it.

The existing API design choice makes it difficult to support future API versions, as other non-Linux hosts do not implement IPSEC offload. If we add support for v1.5 to the Linux PF, then we lose support for IPSEC offload.

A full solution likely requires a new mailbox API with a proper negotiation to check that IPSEC is actually supported by the host.

Fixes 339f28964147 ("ixgbevf: Add support for new mailbox communication between PF and VF")

Signed-off-by Jacob Keller <[email protected]>

Reviewed-by Przemek Kitszel <[email protected]>

Tested-by Rafal Romanowski <[email protected]>

Signed-off-by Tony Nguyen <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/intel/ixgbevf/ipsec.c | 1 -
 1 file changed, 1 deletion(-)

drivers/net/ethernet/intel/ixgbevf/ipsec.c +0 -1

index 66cf17f194082..f804b35d79c72 100644
@@ -629,7 +629,6 @@ void ixgbevf_init_ipsec_offload(struct ixgbevf_adapter *adapter)
 
 	switch (adapter->hw.api_version) {
 	case ixgbe_mbox_api_14:
-	case ixgbe_mbox_api_15:
 		break;
 	default:
 		return;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #47 lore raw

PATCH 47/356 ixgbe: downgrade logging of unsupported VF API version to debug

Link https://lore.kernel.org/intel-wired-lan/[email protected]/ open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jacob Keller <[email protected]>

[ Upstream commit 15915b43a7fb938934bb7fc4290127218859d795 ]

The ixgbe PF driver logs an info message when a VF attempts to negotiate an API version which it does not support:

VF 0 requested invalid api version 6

The ixgbevf driver attempts to load with mailbox API v1.5, which is required for best compatibility with other hosts such as the ESX VMWare PF.

The Linux PF only supports API v1.4, and does not currently have support for the v1.5 API.

The logged message can confuse users, as the v1.5 API is valid, but just happens to not currently be supported by the Linux PF.

Downgrade the info message to a debug message, and fix the language to use 'unsupported' instead of 'invalid' to improve message clarity.

Long term, we should investigate whether the improvements in the v1.5 API make sense for the Linux PF, and if so implement them properly. This may require yet another API version to resolve issues with negotiating IPSEC offload support.

Fixes 339f28964147 ("ixgbevf: Add support for new mailbox communication between PF and VF")

Reported-by Yifei Liu <[email protected]>

Signed-off-by Jacob Keller <[email protected]>

Reviewed-by Przemek Kitszel <[email protected]>

Tested-by Rafal Romanowski <[email protected]>

Signed-off-by Tony Nguyen <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/intel/ixgbe/ixgbe_common.h | 2 ++
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c  | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

drivers/net/ethernet/intel/ixgbe/ixgbe_common.h +2 -0

index 34761e691d52d..efdc222e183d1 100644
@@ -194,6 +194,8 @@ u32 ixgbe_read_reg(struct ixgbe_hw *hw, u32 reg);
 	dev_err(&adapter->pdev->dev, format, ## arg)
 #define e_dev_notice(format, arg...) \
 	dev_notice(&adapter->pdev->dev, format, ## arg)
+#define e_dbg(msglvl, format, arg...) \
+	netif_dbg(adapter, msglvl, adapter->netdev, format, ## arg)
 #define e_info(msglvl, format, arg...) \
 	netif_info(adapter, msglvl, adapter->netdev, format, ## arg)
 #define e_err(msglvl, format, arg...) \

drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c +1 -1

index d0a6c220a12ac..9c89a87e35e01 100644
@@ -1049,7 +1049,7 @@ static int ixgbe_negotiate_vf_api(struct ixgbe_adapter *adapter,
 		break;
 	}
 
-	e_info(drv, "VF %d requested invalid api version %u\n", vf, api);
+	e_dbg(drv, "VF %d requested unsupported api version %u\n", vf, api);
 
 	return -1;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #48 lore raw

PATCH 48/356 igb: Fix potential invalid memory access in igb_init_module()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yuan Can <[email protected]>

[ Upstream commit 0566f83d206c7a864abcd741fe39d6e0ae5eef29 ]

The pci_register_driver() can fail and when this happened, the dca_notifier needs to be unregistered, otherwise the dca_notifier can be called when igb fails to install, resulting to invalid memory access.

Fixes bbd98fe48a43 ("igb: Fix DCA errors and do not use context index for 82576")

Signed-off-by Yuan Can <[email protected]>

Tested-by Pucha Himasekhar Reddy <[email protected]> (A Contingent worker at Intel)

Signed-off-by Tony Nguyen <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/intel/igb/igb_main.c | 4 ++++
 1 file changed, 4 insertions(+)

drivers/net/ethernet/intel/igb/igb_main.c +4 -0

index 49b349fa22542..c38be2880efcf 100644
@@ -665,6 +665,10 @@ static int __init igb_init_module(void)
 	dca_register_notify(&dca_notifier);
 #endif
 	ret = pci_register_driver(&igb_driver);
+#ifdef CONFIG_IGB_DCA
+	if (ret)
+		dca_unregister_notify(&dca_notifier);
+#endif
 	return ret;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #49 lore raw

PATCH 49/356 netfilter: nft_inner: incorrect percpu area handling under softirq

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 7b1d83da254be3bf054965c8f3b1ad976f460ae5 ]

Softirq can interrupt ongoing packet from process context that is walking over the percpu area that contains inner header offsets.

Disable bh and perform three checks before restoring the percpu inner header offsets to validate that the percpu area is valid for this skbuff:

1) If the NFT_PKTINFO_INNER_FULL flag is set on, then this skbuff has already been parsed before for inner header fetching to register.

2) Validate that the percpu area refers to this skbuff using the skbuff pointer as a cookie. If there is a cookie mismatch, then this skbuff needs to be parsed again.

3) Finally, validate if the percpu area refers to this tunnel type.

Only after these three checks the percpu area is restored to a on-stack copy and bh is enabled again.

After inner header fetching, the on-stack copy is stored back to the percpu area.

Fixes 3a07327d10a0 ("netfilter: nft_inner: support for inner tunnel header matching")

Reported-by [email protected]

Signed-off-by Pablo Neira Ayuso <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/netfilter/nf_tables_core.h |  1 +
 net/netfilter/nft_inner.c              | 57 ++++++++++++++++++++------
 2 files changed, 46 insertions(+), 12 deletions(-)

include/net/netfilter/nf_tables_core.h +1 -0

index 780a5f6ad4a67..16855c2a03f8e 100644
@@ -161,6 +161,7 @@ enum {
 };
 
 struct nft_inner_tun_ctx {
+	unsigned long cookie;
 	u16	type;
 	u16	inner_tunoff;
 	u16	inner_lloff;

net/netfilter/nft_inner.c +45 ~~-12~~

107 lines not on this page · raw message

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #50 lore raw

PATCH 50/356 net: sched: fix erspan_opt settings in cls_flower

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xin Long <[email protected]>

[ Upstream commit 292207809486d99c78068d3f459cbbbffde88415 ]

When matching erspan_opt in cls_flower, only the (version, dir, hwid) fields are relevant. However, in fl_set_erspan_opt() it initializes all bits of erspan_opt and its mask to 1. This inadvertently requires packets to match not only the (version, dir, hwid) fields but also the other fields that are unexpectedly set to 1.

This patch resolves the issue by ensuring that only the (version, dir, hwid) fields are configured in fl_set_erspan_opt(), leaving the other fields to 0 in erspan_opt.

Fixes 79b1011cb33d ("net: sched: allow flower to match erspan options")

Reported-by Shuang Li <[email protected]>

Signed-off-by Xin Long <[email protected]>

Reviewed-by Cong Wang <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/sched/cls_flower.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

net/sched/cls_flower.c +3 -2

index 6ee7064c82fcc..bcf1b8012b2c3 100644
@@ -1320,7 +1320,6 @@ static int fl_set_erspan_opt(const struct nlattr *nla, struct fl_flow_key *key,
 	int err;
 
 	md = (struct erspan_metadata *)&key->enc_opts.data[key->enc_opts.len];
-	memset(md, 0xff, sizeof(*md));
 	md->version = 1;
 
 	if (!depth)

@@ -1349,9 +1348,9 @@ static int fl_set_erspan_opt(const struct nlattr *nla, struct fl_flow_key *key,
 			NL_SET_ERR_MSG(extack, "Missing tunnel key erspan option index");
 			return -EINVAL;
 		}
+		memset(&md->u.index, 0xff, sizeof(md->u.index));
 		if (tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_INDEX]) {
 			nla = tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_INDEX];
-			memset(&md->u, 0x00, sizeof(md->u));
 			md->u.index = nla_get_be32(nla);
 		}
 	} else if (md->version == 2) {

@@ -1360,10 +1359,12 @@ static int fl_set_erspan_opt(const struct nlattr *nla, struct fl_flow_key *key,
 			NL_SET_ERR_MSG(extack, "Missing tunnel key erspan option dir or hwid");
 			return -EINVAL;
 		}
+		md->u.md2.dir = 1;
 		if (tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_DIR]) {
 			nla = tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_DIR];
 			md->u.md2.dir = nla_get_u8(nla);
 		}
+		set_hwid(&md->u.md2, 0xff);
 		if (tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_HWID]) {
 			nla = tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_HWID];
 			set_hwid(&md->u.md2, nla_get_u8(nla));

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #51 lore raw

PATCH 51/356 netfilter: ipset: Hold module reference while requesting a module

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Phil Sutter <[email protected]>

[ Upstream commit 456f010bfaefde84d3390c755eedb1b0a5857c3c ]

User space may unload ip_set.ko while it is itself requesting a set type backend module, leading to a kernel crash. The race condition may be provoked by inserting an mdelay() right after the nfnl_unlock() call.

Fixes a7b4f989a629 ("netfilter: ipset: IP set core support")

Signed-off-by Phil Sutter <[email protected]>

Acked-by Jozsef Kadlecsik <[email protected]>

Signed-off-by Pablo Neira Ayuso <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/netfilter/ipset/ip_set_core.c | 5 +++++
 1 file changed, 5 insertions(+)

net/netfilter/ipset/ip_set_core.c +5 -0

index 61431690cbd5f..cc20e6d56807c 100644
@@ -104,14 +104,19 @@ find_set_type(const char *name, u8 family, u8 revision)
 static bool
 load_settype(const char *name)
 {
+	if (!try_module_get(THIS_MODULE))
+		return false;
+
 	nfnl_unlock(NFNL_SUBSYS_IPSET);
 	pr_debug("try to load ip_set_%s\n", name);
 	if (request_module("ip_set_%s", name) < 0) {
 		pr_warn("Can't find ip_set type %s\n", name);
 		nfnl_lock(NFNL_SUBSYS_IPSET);
+		module_put(THIS_MODULE);
 		return false;
 	}
 	nfnl_lock(NFNL_SUBSYS_IPSET);
+	module_put(THIS_MODULE);
 	return true;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #52 lore raw

PATCH 52/356 netfilter: nft_set_hash: skip duplicated elements pending gc run

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 7ffc7481153bbabf3332c6a19b289730c7e1edf5 ]

rhashtable does not provide stable walk, duplicated elements are possible in case of resizing. I considered that checking for errors when calling rhashtable_walk_next() was sufficient to detect the resizing. However, rhashtable_walk_next() returns -EAGAIN only at the end of the iteration, which is too late, because a gc work containing duplicated elements could have been already scheduled for removal to the worker.

Add a u32 gc worker sequence number per set, bump it on every workqueue run. Annotate gc worker sequence number on the expired element. Use it to skip those already seen in this gc workqueue run.

Note that this new field is never reset in case gc transaction fails, so next gc worker run on the expired element overrides it. Wraparound of gc worker sequence number should not be an issue with stale gc worker sequence number in the element, that would just postpone the element removal in one gc run.

Note that it is not possible to use flags to annotate that element is pending gc run to detect duplicates, given that gc transaction can be invalidated in case of update from the control plane, therefore, not allowing to clear such flag.

On x86_64, pahole reports no changes in the size of nft_rhash_elem.

Fixes f6c383b8c31a ("netfilter: nf_tables: adapt set backend to use GC transaction API")

Reported-by Laurent Fasnacht <[email protected]>

Tested-by Laurent Fasnacht <[email protected]>

Signed-off-by Pablo Neira Ayuso <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/netfilter/nft_set_hash.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

net/netfilter/nft_set_hash.c +16 -0

index 3a96d4a77a228..cc1ae18485faf 100644
@@ -24,10 +24,12 @@
 struct nft_rhash {
 	struct rhashtable		ht;
 	struct delayed_work		gc_work;
+	u32				wq_gc_seq;
 };
 
 struct nft_rhash_elem {
 	struct rhash_head		node;
+	u32				wq_gc_seq;
 	struct nft_set_ext		ext;
 };

@@ -331,6 +333,10 @@ static void nft_rhash_gc(struct work_struct *work)
 	if (!gc)
 		goto done;
 
+	/* Elements never collected use a zero gc worker sequence number. */
+	if (unlikely(++priv->wq_gc_seq == 0))
+		priv->wq_gc_seq++;
+
 	rhashtable_walk_enter(&priv->ht, &hti);
 	rhashtable_walk_start(&hti);

@@ -348,6 +354,14 @@ static void nft_rhash_gc(struct work_struct *work)
 			goto try_later;
 		}
 
+		/* rhashtable walk is unstable, already seen in this gc run?
+		 * Then, skip this element. In case of (unlikely) sequence
+		 * wraparound and stale element wq_gc_seq, next gc run will
+		 * just find this expired element.
+		 */
+		if (he->wq_gc_seq == priv->wq_gc_seq)
+			continue;
+
 		if (nft_set_elem_is_dead(&he->ext))
 			goto dead_elem;

@@ -364,6 +378,8 @@ static void nft_rhash_gc(struct work_struct *work)
 		if (!gc)
 			goto try_later;
 
+		/* annotate gc sequence for this attempt. */
+		he->wq_gc_seq = priv->wq_gc_seq;
 		nft_trans_gc_elem_add(gc, he);
 	}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:32 UTC #53 lore raw

PATCH 53/356 ethtool: Fix wrong mod state in case of verbose and no_mask bitset

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kory Maincent <[email protected]>

[ Upstream commit 910c4788d6155b2202ec88273376cd7ecdc24f0a ]

A bitset without mask in a _SET request means we want exactly the bits in the bitset to be set. This works correctly for compact format but when verbose format is parsed, ethnl_update_bitset32_verbose() only sets the bits present in the request bitset but does not clear the rest. The commit 6699170376ab ("ethtool: fix application of verbose no_mask bitset") fixes this issue by clearing the whole target bitmap before we start iterating. The solution proposed brought an issue with the behavior of the mod variable. As the bitset is always cleared the old value will always differ to the new value.

Fix it by adding a new function to compare bitmaps and a temporary variable which save the state of the old bitmap.

Fixes 6699170376ab ("ethtool: fix application of verbose no_mask bitset")

Signed-off-by Kory Maincent <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 net/ethtool/bitset.c | 48 ++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 44 insertions(+), 4 deletions(-)

net/ethtool/bitset.c +44 -4

index 0515d6604b3b9..f0883357d12e5 100644
@@ -425,12 +425,32 @@ static int ethnl_parse_bit(unsigned int *index, bool *val, unsigned int nbits,
 	return 0;
 }
 
+/**
+ * ethnl_bitmap32_equal() - Compare two bitmaps
+ * @map1:  first bitmap
+ * @map2:  second bitmap
+ * @nbits: bit size to compare
+ *
+ * Return: true if first @nbits are equal, false if not
+ */
+static bool ethnl_bitmap32_equal(const u32 *map1, const u32 *map2,
+				 unsigned int nbits)
+{
+	if (memcmp(map1, map2, nbits / 32 * sizeof(u32)))
+		return false;
+	if (nbits % 32 == 0)
+		return true;
+	return !((map1[nbits / 32] ^ map2[nbits / 32]) &
+		 ethnl_lower_bits(nbits % 32));
+}
+
 static int
 ethnl_update_bitset32_verbose(u32 *bitmap, unsigned int nbits,
 			      const struct nlattr *attr, struct nlattr **tb,
 			      ethnl_string_array_t names,
 			      struct netlink_ext_ack *extack, bool *mod)
 {
+	u32 *saved_bitmap = NULL;
 	struct nlattr *bit_attr;
 	bool no_mask;
 	int rem;

@@ -448,8 +468,20 @@ ethnl_update_bitset32_verbose(u32 *bitmap, unsigned int nbits,
 	}
 
 	no_mask = tb[ETHTOOL_A_BITSET_NOMASK];
-	if (no_mask)
-		ethnl_bitmap32_clear(bitmap, 0, nbits, mod);
+	if (no_mask) {
+		unsigned int nwords = DIV_ROUND_UP(nbits, 32);
+		unsigned int nbytes = nwords * sizeof(u32);
+		bool dummy;
+
+		/* The bitmap size is only the size of the map part without
+		 * its mask part.
+		 */
+		saved_bitmap = kcalloc(nwords, sizeof(u32), GFP_KERNEL);
+		if (!saved_bitmap)
+			return -ENOMEM;
+		memcpy(saved_bitmap, bitmap, nbytes);
+		ethnl_bitmap32_clear(bitmap, 0, nbits, &dummy);
+	}
 
 	nla_for_each_nested(bit_attr, tb[ETHTOOL_A_BITSET_BITS], rem) {
 		bool old_val, new_val;

@@ -458,22 +490,30 @@ ethnl_update_bitset32_verbose(u32 *bitmap, unsigned int nbits,
 		if (nla_type(bit_attr) != ETHTOOL_A_BITSET_BITS_BIT) {
 			NL_SET_ERR_MSG_ATTR(extack, bit_attr,
 					    "only ETHTOOL_A_BITSET_BITS_BIT allowed in ETHTOOL_A_BITSET_BITS");
+			kfree(saved_bitmap);
 			return -EINVAL;
 		}
 		ret = ethnl_parse_bit(&idx, &new_val, nbits, bit_attr, no_mask,
 				      names, extack);
-		if (ret < 0)
+		if (ret < 0) {
+			kfree(saved_bitmap);
 			return ret;
+		}
 		old_val = bitmap[idx / 32] & ((u32)1 << (idx % 32));
 		if (new_val != old_val) {
 			if (new_val)
 				bitmap[idx / 32] |= ((u32)1 << (idx % 32));
 			else
 				bitmap[idx / 32] &= ~((u32)1 << (idx % 32));
-			*mod = true;
+			if (!no_mask)
+				*mod = true;
 		}
 	}
 
+	if (no_mask && !ethnl_bitmap32_equal(saved_bitmap, bitmap, nbits))
+		*mod = true;
+
+	kfree(saved_bitmap);
 	return 0;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #54 lore raw

PATCH 54/356 mlxsw: Add ipv4_5 flex key

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Amit Cohen <[email protected]>

[ Upstream commit c2f3e10ac4ebf23e177226b9b4d297bfe2fb6b20 ]

Currently virtual router ID element is broken to two sub-elements - 'VIRT_ROUTER_LSB' and 'VIRT_ROUTER_MSB'. It was broken as this field is broken in 'ipv4_4' flex key which is used for IPv4 in Spectrum < 4. For Spectrum-4, we use 'ipv4_4b' flex key which contains one field for virtual router, this key is not supported in older ASICs.

Add 'ipv4_5' flex key which is supported in all ASICs and contains one field for virtual router. Then there is no reason to use 'VIRT_ROUTER_LSB' and 'VIRT_ROUTER_MSB', remove them and add one element 'VIRT_ROUTER' for this field.

The motivation is to get rid of 'ipv4_4' flex key, as it might be chosen for IPv6 multicast forwarding region. This will not allow the improvement in a following patch. See more details in the cover letter and in a following patch.

Signed-off-by Amit Cohen <[email protected]>

Stable-dep-of: 217bbf156f93 ("mlxsw: spectrum_acl_flex_keys: Use correct key block on Spectrum-4")

Signed-off-by Sasha Levin <[email protected]>

---
 .../ethernet/mellanox/mlxsw/core_acl_flex_keys.c    |  3 +--
 .../ethernet/mellanox/mlxsw/core_acl_flex_keys.h    |  3 +--
 .../net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c | 13 ++++---------
 .../mellanox/mlxsw/spectrum_acl_flex_keys.c         | 10 ++++------
 4 files changed, 10 insertions(+), 19 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c +1 -2

index bf140e7416e19..5fa3800940c89 100644
@@ -33,8 +33,7 @@ static const struct mlxsw_afk_element_info mlxsw_afk_element_infos[] = {
 	MLXSW_AFK_ELEMENT_INFO_U32(IP_TTL_, 0x18, 0, 8),
 	MLXSW_AFK_ELEMENT_INFO_U32(IP_ECN, 0x18, 9, 2),
 	MLXSW_AFK_ELEMENT_INFO_U32(IP_DSCP, 0x18, 11, 6),
-	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_MSB, 0x18, 17, 4),
-	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_LSB, 0x18, 21, 8),
+	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER, 0x18, 17, 12),
 	MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_96_127, 0x20, 4),
 	MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_64_95, 0x24, 4),
 	MLXSW_AFK_ELEMENT_INFO_BUF(SRC_IP_32_63, 0x28, 4),

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h +1 -2

index 2eac7582c31a8..75e9bbc361701 100644
@@ -33,8 +33,7 @@ enum mlxsw_afk_element {
 	MLXSW_AFK_ELEMENT_IP_TTL_,
 	MLXSW_AFK_ELEMENT_IP_ECN,
 	MLXSW_AFK_ELEMENT_IP_DSCP,
-	MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
-	MLXSW_AFK_ELEMENT_VIRT_ROUTER_LSB,
+	MLXSW_AFK_ELEMENT_VIRT_ROUTER,
 	MLXSW_AFK_ELEMENT_FDB_MISS,
 	MLXSW_AFK_ELEMENT_L4_PORT_RANGE,
 	MLXSW_AFK_ELEMENT_MAX,

drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c +4 -9

index b1178b7a7f51a..2efcc9372d4e6 100644
@@ -45,8 +45,7 @@ static int mlxsw_sp2_mr_tcam_bind_group(struct mlxsw_sp *mlxsw_sp,
 }
 
 static const enum mlxsw_afk_element mlxsw_sp2_mr_tcam_usage_ipv4[] = {
-		MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
-		MLXSW_AFK_ELEMENT_VIRT_ROUTER_LSB,
+		MLXSW_AFK_ELEMENT_VIRT_ROUTER,
 		MLXSW_AFK_ELEMENT_SRC_IP_0_31,
 		MLXSW_AFK_ELEMENT_DST_IP_0_31,
 };

@@ -89,8 +88,7 @@ static void mlxsw_sp2_mr_tcam_ipv4_fini(struct mlxsw_sp2_mr_tcam *mr_tcam)
 }
 
 static const enum mlxsw_afk_element mlxsw_sp2_mr_tcam_usage_ipv6[] = {
-		MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
-		MLXSW_AFK_ELEMENT_VIRT_ROUTER_LSB,
+		MLXSW_AFK_ELEMENT_VIRT_ROUTER,
 		MLXSW_AFK_ELEMENT_SRC_IP_96_127,
 		MLXSW_AFK_ELEMENT_SRC_IP_64_95,
 		MLXSW_AFK_ELEMENT_SRC_IP_32_63,

@@ -189,11 +187,8 @@ mlxsw_sp2_mr_tcam_rule_parse(struct mlxsw_sp_acl_rule *rule,
 
 	rulei = mlxsw_sp_acl_rule_rulei(rule);
 	rulei->priority = priority;
-	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER_LSB,
-				       key->vrid, GENMASK(7, 0));
-	mlxsw_sp_acl_rulei_keymask_u32(rulei,
-				       MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
-				       key->vrid >> 8, GENMASK(3, 0));
+	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER,
+				       key->vrid, GENMASK(11, 0));
 	switch (key->proto) {
 	case MLXSW_SP_L3_PROTO_IPV4:
 		return mlxsw_sp2_mr_tcam_rule_parse4(rulei, key);

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c +4 -6

index cb746a43b24b3..cc00c8d69eb77 100644
@@ -171,9 +171,8 @@ static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_2[] = {
 	MLXSW_AFK_ELEMENT_INST_U32(IP_PROTO, 0x04, 16, 8),
 };
 
-static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_4[] = {
-	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_LSB, 0x04, 24, 8),
-	MLXSW_AFK_ELEMENT_INST_EXT_U32(VIRT_ROUTER_MSB, 0x00, 0, 3, 0, true),
+static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_5[] = {
+	MLXSW_AFK_ELEMENT_INST_EXT_U32(VIRT_ROUTER, 0x04, 20, 11, 0, true),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_0[] = {

@@ -220,7 +219,7 @@ static const struct mlxsw_afk_block mlxsw_sp2_afk_blocks[] = {
 	MLXSW_AFK_BLOCK(0x38, mlxsw_sp_afk_element_info_ipv4_0),
 	MLXSW_AFK_BLOCK(0x39, mlxsw_sp_afk_element_info_ipv4_1),
 	MLXSW_AFK_BLOCK(0x3A, mlxsw_sp_afk_element_info_ipv4_2),
-	MLXSW_AFK_BLOCK(0x3C, mlxsw_sp_afk_element_info_ipv4_4),
+	MLXSW_AFK_BLOCK(0x3D, mlxsw_sp_afk_element_info_ipv4_5),
 	MLXSW_AFK_BLOCK(0x40, mlxsw_sp_afk_element_info_ipv6_0),
 	MLXSW_AFK_BLOCK(0x41, mlxsw_sp_afk_element_info_ipv6_1),
 	MLXSW_AFK_BLOCK(0x42, mlxsw_sp_afk_element_info_ipv6_2),

@@ -323,8 +322,7 @@ static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_mac_5b[] = {
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_4b[] = {
-	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_LSB, 0x04, 13, 8),
-	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_MSB, 0x04, 21, 4),
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER, 0x04, 13, 12),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_2b[] = {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #55 lore raw

PATCH 55/356 mlxsw: spectrum_acl_flex_keys: Add ipv4_5b flex key

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Amit Cohen <[email protected]>

[ Upstream commit c6caabdf3e0cc50ba4a44bebc82cda5551d81d4f ]

The previous patch replaced the key block 'ipv4_4' with 'ipv4_5'. The corresponding block for Spectrum-4 is 'ipv4_4b'. To be consistent, replace key block 'ipv4_4b' with 'ipv4_5b'.

Signed-off-by Amit Cohen <[email protected]>

Stable-dep-of: 217bbf156f93 ("mlxsw: spectrum_acl_flex_keys: Use correct key block on Spectrum-4")

Signed-off-by Sasha Levin <[email protected]>

---
 .../net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c    | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c +3 -3

index cc00c8d69eb77..7d66c4f2deeaa 100644
@@ -321,8 +321,8 @@ static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_mac_5b[] = {
 	MLXSW_AFK_ELEMENT_INST_EXT_U32(SRC_SYS_PORT, 0x04, 0, 9, -1, true), /* RX_ACL_SYSTEM_PORT */
 };
 
-static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_4b[] = {
-	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER, 0x04, 13, 12),
+static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_5b[] = {
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER, 0x04, 20, 12),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_2b[] = {

@@ -339,7 +339,7 @@ static const struct mlxsw_afk_block mlxsw_sp4_afk_blocks[] = {
 	MLXSW_AFK_BLOCK(0x38, mlxsw_sp_afk_element_info_ipv4_0),
 	MLXSW_AFK_BLOCK(0x39, mlxsw_sp_afk_element_info_ipv4_1),
 	MLXSW_AFK_BLOCK(0x3A, mlxsw_sp_afk_element_info_ipv4_2),
-	MLXSW_AFK_BLOCK(0x35, mlxsw_sp_afk_element_info_ipv4_4b),
+	MLXSW_AFK_BLOCK(0x36, mlxsw_sp_afk_element_info_ipv4_5b),
 	MLXSW_AFK_BLOCK(0x40, mlxsw_sp_afk_element_info_ipv6_0),
 	MLXSW_AFK_BLOCK(0x41, mlxsw_sp_afk_element_info_ipv6_1),
 	MLXSW_AFK_BLOCK(0x47, mlxsw_sp_afk_element_info_ipv6_2b),

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #56 lore raw

PATCH 56/356 mlxsw: Edit IPv6 key blocks to use one less block for multicast forwarding

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Amit Cohen <[email protected]>

[ Upstream commit 92953e7aab013719aa8974805614c0bc11361026 ]

Two ACL regions that are configured by the driver during initialization are the ones used for IPv4 and IPv6 multicast forwarding. Entries residing in these two regions match on the {SIP, DIP, VRID} key elements.

Currently for IPv6 region, 9 key blocks are used:
* 4 for SIP - 'ipv4_1', 'ipv6_{3,4,5}'
* 4 for DIP - 'ipv4_0', 'ipv6_{0,1,2/2b}'
* 1 for VRID - 'ipv4_4b'

This can be improved by reducing the amount key blocks needed for the IPv6 region to 8. It is possible to use key blocks that mix subsets of the VRID element with subsets of the DIP element.
The following key blocks can be used:
* 4 for SIP - 'ipv4_1', 'ipv6_{3,4,5}'
* 1 for subset of DIP - 'ipv4_0'
* 3 for the rest of DIP and subsets of VRID - 'ipv6_{0,1,2/2b}'

To make this happen, add VRID sub-elements as part of existing keys - 'ipv6_{0,1,2/2b}'. Note that one of the sub-elements is called VRID_ROUTER_MSB and does not contain bit numbers like the rest, as for Spectrum < 4 this element represents bits 8-10 and for Spectrum-4 it represents bits 8-11.

Breaking VRID into 3 sub-elements makes the driver use one less block in IPv6 region for multicast forwarding. The sub-elements can be filled in blocks that are used for destination IP.

The algorithm in the driver that chooses which key blocks will be used is lazy and not the optimal one. It searches the block that contains the most elements that are required, chooses it, removes the elements that appear in the chosen block and starts again searching the block that contains the most elements.

When key block 'ipv4_4' is defined, the algorithm might choose it, as it contains 2 sub-elements of VRID, then 8 blocks must be chosen for SIP and DIP and we get 9 blocks to match on {SIP, DIP, VRID}. That is why we had to remove key block 'ipv4_4' in a previous patch and use key block that contains one field for VRID.

This improvement was tested and indeed 8 blocks are used instead of 9.

Signed-off-by Amit Cohen <[email protected]>

Stable-dep-of: 217bbf156f93 ("mlxsw: spectrum_acl_flex_keys: Use correct key block on Spectrum-4")

Signed-off-by Sasha Levin <[email protected]>

---
 .../ethernet/mellanox/mlxsw/core_acl_flex_keys.c  |  3 +++
 .../ethernet/mellanox/mlxsw/core_acl_flex_keys.h  |  3 +++
 .../ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c   | 15 ++++++++++++---
 .../mellanox/mlxsw/spectrum_acl_flex_keys.c       |  4 ++++
 4 files changed, 22 insertions(+), 3 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c +3 -0

index 5fa3800940c89..654dafc9b54d3 100644
@@ -44,6 +44,9 @@ static const struct mlxsw_afk_element_info mlxsw_afk_element_infos[] = {
 	MLXSW_AFK_ELEMENT_INFO_BUF(DST_IP_0_31, 0x3C, 4),
 	MLXSW_AFK_ELEMENT_INFO_U32(FDB_MISS, 0x40, 0, 1),
 	MLXSW_AFK_ELEMENT_INFO_U32(L4_PORT_RANGE, 0x40, 1, 16),
+	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_0_3, 0x40, 17, 4),
+	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_4_7, 0x40, 21, 4),
+	MLXSW_AFK_ELEMENT_INFO_U32(VIRT_ROUTER_MSB, 0x40, 25, 4),
 };
 
 struct mlxsw_afk {

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h +3 -0

index 75e9bbc361701..1c76aa3ffab72 100644
@@ -36,6 +36,9 @@ enum mlxsw_afk_element {
 	MLXSW_AFK_ELEMENT_VIRT_ROUTER,
 	MLXSW_AFK_ELEMENT_FDB_MISS,
 	MLXSW_AFK_ELEMENT_L4_PORT_RANGE,
+	MLXSW_AFK_ELEMENT_VIRT_ROUTER_0_3,
+	MLXSW_AFK_ELEMENT_VIRT_ROUTER_4_7,
+	MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
 	MLXSW_AFK_ELEMENT_MAX,
 };

drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c +12 -3

index 2efcc9372d4e6..99eeafdc8d1e4 100644
@@ -88,7 +88,9 @@ static void mlxsw_sp2_mr_tcam_ipv4_fini(struct mlxsw_sp2_mr_tcam *mr_tcam)
 }
 
 static const enum mlxsw_afk_element mlxsw_sp2_mr_tcam_usage_ipv6[] = {
-		MLXSW_AFK_ELEMENT_VIRT_ROUTER,
+		MLXSW_AFK_ELEMENT_VIRT_ROUTER_0_3,
+		MLXSW_AFK_ELEMENT_VIRT_ROUTER_4_7,
+		MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
 		MLXSW_AFK_ELEMENT_SRC_IP_96_127,
 		MLXSW_AFK_ELEMENT_SRC_IP_64_95,
 		MLXSW_AFK_ELEMENT_SRC_IP_32_63,

@@ -140,6 +142,8 @@ static void
 mlxsw_sp2_mr_tcam_rule_parse4(struct mlxsw_sp_acl_rule_info *rulei,
 			      struct mlxsw_sp_mr_route_key *key)
 {
+	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER,
+				       key->vrid, GENMASK(11, 0));
 	mlxsw_sp_acl_rulei_keymask_buf(rulei, MLXSW_AFK_ELEMENT_SRC_IP_0_31,
 				       (char *) &key->source.addr4,
 				       (char *) &key->source_mask.addr4, 4);

@@ -152,6 +156,13 @@ static void
 mlxsw_sp2_mr_tcam_rule_parse6(struct mlxsw_sp_acl_rule_info *rulei,
 			      struct mlxsw_sp_mr_route_key *key)
 {
+	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER_0_3,
+				       key->vrid, GENMASK(3, 0));
+	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER_4_7,
+				       key->vrid >> 4, GENMASK(3, 0));
+	mlxsw_sp_acl_rulei_keymask_u32(rulei,
+				       MLXSW_AFK_ELEMENT_VIRT_ROUTER_MSB,
+				       key->vrid >> 8, GENMASK(3, 0));
 	mlxsw_sp_acl_rulei_keymask_buf(rulei, MLXSW_AFK_ELEMENT_SRC_IP_96_127,
 				       &key->source.addr6.s6_addr[0x0],
 				       &key->source_mask.addr6.s6_addr[0x0], 4);

@@ -187,8 +198,6 @@ mlxsw_sp2_mr_tcam_rule_parse(struct mlxsw_sp_acl_rule *rule,
 
 	rulei = mlxsw_sp_acl_rule_rulei(rule);
 	rulei->priority = priority;
-	mlxsw_sp_acl_rulei_keymask_u32(rulei, MLXSW_AFK_ELEMENT_VIRT_ROUTER,
-				       key->vrid, GENMASK(11, 0));
 	switch (key->proto) {
 	case MLXSW_SP_L3_PROTO_IPV4:
 		return mlxsw_sp2_mr_tcam_rule_parse4(rulei, key);

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c +4 -0

index 7d66c4f2deeaa..4b3564f5fd652 100644
@@ -176,14 +176,17 @@ static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_5[] = {
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_0[] = {
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_0_3, 0x00, 0, 4),
 	MLXSW_AFK_ELEMENT_INST_BUF(DST_IP_32_63, 0x04, 4),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_1[] = {
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_4_7, 0x00, 0, 4),
 	MLXSW_AFK_ELEMENT_INST_BUF(DST_IP_64_95, 0x04, 4),
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_2[] = {
+	MLXSW_AFK_ELEMENT_INST_EXT_U32(VIRT_ROUTER_MSB, 0x00, 0, 3, 0, true),
 	MLXSW_AFK_ELEMENT_INST_BUF(DST_IP_96_127, 0x04, 4),
 };

@@ -326,6 +329,7 @@ static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_5b[] = {
 };
 
 static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_2b[] = {
+	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER_MSB, 0x00, 0, 4),
 	MLXSW_AFK_ELEMENT_INST_BUF(DST_IP_96_127, 0x04, 4),
 };

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #57 lore raw

PATCH 57/356 mlxsw: Mark high entropy key blocks

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Amit Cohen <[email protected]>

[ Upstream commit cad6431b867507779c41b00baaf18382467ef0a0 ]

For 12 key blocks in the A-TCAM, rules are split into two records, which constitute two lookups. The two records are linked using a "large entry key ID".

Due to a Spectrum-4 hardware issue, KVD entries that correspond to key blocks 0 to 5 of 12 key blocks A-TCAM entries will be placed in the same KVD pipe if they only differ in their "large entry key ID", as it is ignored. This results in a reduced scale. To reduce the probability of this issue, we can place key blocks with high entropy in blocks 0 to 5. The idea is to place blocks that are changed often in blocks 0 to 5, for example, key blocks that match on IPv4 addresses or the LSBs of IPv6 addresses. Such placement will reduce the probability of these blocks to be same.

Mark several blocks with 'high_entropy' flag, so later we will take into account this flag and place them in blocks 0 to 5.

Signed-off-by Amit Cohen <[email protected]>

Stable-dep-of: 217bbf156f93 ("mlxsw: spectrum_acl_flex_keys: Use correct key block on Spectrum-4")

Signed-off-by Sasha Levin <[email protected]>

---
 .../net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h |  9 +++++++++
 .../ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c | 12 ++++++------
 2 files changed, 15 insertions(+), 6 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h +9 -0

index 1c76aa3ffab72..98a05598178b3 100644
@@ -119,6 +119,7 @@ struct mlxsw_afk_block {
 	u16 encoding; /* block ID */
 	struct mlxsw_afk_element_inst *instances;
 	unsigned int instances_count;
+	bool high_entropy;
 };
 
 #define MLXSW_AFK_BLOCK(_encoding, _instances)					\

@@ -128,6 +129,14 @@ struct mlxsw_afk_block {
 		.instances_count = ARRAY_SIZE(_instances),			\
 	}
 
+#define MLXSW_AFK_BLOCK_HIGH_ENTROPY(_encoding, _instances)			\
+	{									\
+		.encoding = _encoding,						\
+		.instances = _instances,					\
+		.instances_count = ARRAY_SIZE(_instances),			\
+		.high_entropy = true,						\
+	}
+
 struct mlxsw_afk_element_usage {
 	DECLARE_BITMAP(usage, MLXSW_AFK_ELEMENT_MAX);
 };

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c +6 -6

index 4b3564f5fd652..eaad786056027 100644
@@ -334,14 +334,14 @@ static struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv6_2b[] = {
 };
 
 static const struct mlxsw_afk_block mlxsw_sp4_afk_blocks[] = {
-	MLXSW_AFK_BLOCK(0x10, mlxsw_sp_afk_element_info_mac_0),
-	MLXSW_AFK_BLOCK(0x11, mlxsw_sp_afk_element_info_mac_1),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x10, mlxsw_sp_afk_element_info_mac_0),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x11, mlxsw_sp_afk_element_info_mac_1),
 	MLXSW_AFK_BLOCK(0x12, mlxsw_sp_afk_element_info_mac_2),
 	MLXSW_AFK_BLOCK(0x13, mlxsw_sp_afk_element_info_mac_3),
 	MLXSW_AFK_BLOCK(0x14, mlxsw_sp_afk_element_info_mac_4),
-	MLXSW_AFK_BLOCK(0x1A, mlxsw_sp_afk_element_info_mac_5b),
-	MLXSW_AFK_BLOCK(0x38, mlxsw_sp_afk_element_info_ipv4_0),
-	MLXSW_AFK_BLOCK(0x39, mlxsw_sp_afk_element_info_ipv4_1),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x1A, mlxsw_sp_afk_element_info_mac_5b),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x38, mlxsw_sp_afk_element_info_ipv4_0),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x39, mlxsw_sp_afk_element_info_ipv4_1),
 	MLXSW_AFK_BLOCK(0x3A, mlxsw_sp_afk_element_info_ipv4_2),
 	MLXSW_AFK_BLOCK(0x36, mlxsw_sp_afk_element_info_ipv4_5b),
 	MLXSW_AFK_BLOCK(0x40, mlxsw_sp_afk_element_info_ipv6_0),

@@ -350,7 +350,7 @@ static const struct mlxsw_afk_block mlxsw_sp4_afk_blocks[] = {
 	MLXSW_AFK_BLOCK(0x43, mlxsw_sp_afk_element_info_ipv6_3),
 	MLXSW_AFK_BLOCK(0x44, mlxsw_sp_afk_element_info_ipv6_4),
 	MLXSW_AFK_BLOCK(0x45, mlxsw_sp_afk_element_info_ipv6_5),
-	MLXSW_AFK_BLOCK(0x90, mlxsw_sp_afk_element_info_l4_0),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x90, mlxsw_sp_afk_element_info_l4_0),
 	MLXSW_AFK_BLOCK(0x92, mlxsw_sp_afk_element_info_l4_2),
 };

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #58 lore raw

PATCH 58/356 mlxsw: spectrum_acl_flex_keys: Constify struct mlxsw_afk_element_inst

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christophe JAILLET <[email protected]>

[ Upstream commit bec2a32145d5cc066df29182fa0e5b0d4329b1a1 ]

'struct mlxsw_afk_element_inst' are not modified in these drivers.

Constifying these structures moves some data to a read-only section, so increases overall security.

Update a few functions and struct mlxsw_afk_block accordingly.

On a x86_64, with allmodconfig, as an example:
Before:
======
text data bss dec hex filename
4278 4032 0 8310 2076 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.o

After:
=====
text data bss dec hex filename
7934 352 0 8286 205e drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.o

Signed-off-by Christophe JAILLET <[email protected]>

Reviewed-by Simon Horman <[email protected]>

Link https://patch.msgid.link/8ccfc7bfb2365dcee5b03c81ebe061a927d6da2e.1727541677.git.christophe.jaillet@wanadoo.fr open here

Stable-dep-of: 217bbf156f93 ("mlxsw: spectrum_acl_flex_keys: Use correct key block on Spectrum-4")

Signed-off-by Sasha Levin <[email protected]>

---
 .../mellanox/mlxsw/core_acl_flex_keys.c       |  6 +-
 .../mellanox/mlxsw/core_acl_flex_keys.h       |  2 +-
 .../mellanox/mlxsw/spectrum_acl_flex_keys.c   | 66 +++++++++----------
 3 files changed, 37 insertions(+), 37 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c +3 -3

index 654dafc9b54d3..57a1a55856a51 100644
@@ -67,7 +67,7 @@ static bool mlxsw_afk_blocks_check(struct mlxsw_afk *mlxsw_afk)
 
 		for (j = 0; j < block->instances_count; j++) {
 			const struct mlxsw_afk_element_info *elinfo;
-			struct mlxsw_afk_element_inst *elinst;
+			const struct mlxsw_afk_element_inst *elinst;
 
 			elinst = &block->instances[j];
 			elinfo = &mlxsw_afk_element_infos[elinst->element];

@@ -153,7 +153,7 @@ static void mlxsw_afk_picker_count_hits(struct mlxsw_afk *mlxsw_afk,
 		const struct mlxsw_afk_block *block = &mlxsw_afk->blocks[i];
 
 		for (j = 0; j < block->instances_count; j++) {
-			struct mlxsw_afk_element_inst *elinst;
+			const struct mlxsw_afk_element_inst *elinst;
 
 			elinst = &block->instances[j];
 			if (elinst->element == element) {

@@ -336,7 +336,7 @@ mlxsw_afk_block_elinst_get(const struct mlxsw_afk_block *block,
 	int i;
 
 	for (i = 0; i < block->instances_count; i++) {
-		struct mlxsw_afk_element_inst *elinst;
+		const struct mlxsw_afk_element_inst *elinst;
 
 		elinst = &block->instances[i];
 		if (elinst->element == element)

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.h +1 -1

index 98a05598178b3..5aa1afb3f2ca8 100644
@@ -117,7 +117,7 @@ struct mlxsw_afk_element_inst { /* element instance in actual block */
 
 struct mlxsw_afk_block {
 	u16 encoding; /* block ID */
-	struct mlxsw_afk_element_inst *instances;
+	const struct mlxsw_afk_element_inst *instances;
 	unsigned int instances_count;
 	bool high_entropy;
 };

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c +33 ~~-33~~

215 lines not on this page · raw message

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #59 lore raw

PATCH 59/356 mlxsw: spectrum_acl_flex_keys: Use correct key block on Spectrum-4

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 217bbf156f93ada86b91617489e7ba8a0904233c ]

The driver is currently using an ACL key block that is not supported by Spectrum-4. This works because the driver is only using a single field from this key block which is located in the same offset in the equivalent Spectrum-4 key block.

The issue was discovered when the firmware started rejecting the use of the unsupported key block. The change has been reverted to avoid breaking users that only update their firmware.

Nonetheless, fix the issue by using the correct key block.

Fixes 07ff135958dd ("mlxsw: Introduce flex key elements for Spectrum-4")

Signed-off-by Ido Schimmel <[email protected]>

Reviewed-by Petr Machata <[email protected]>

Link https://patch.msgid.link/35e72c97bdd3bc414fb8e4d747e5fb5d26c29658.1733237440.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 .../net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c    | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_keys.c +5 -1

index 6fe185ea6732c..1850a975b3804 100644
@@ -324,6 +324,10 @@ static const struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_mac_5b[] =
 	MLXSW_AFK_ELEMENT_INST_EXT_U32(SRC_SYS_PORT, 0x04, 0, 9, -1, true), /* RX_ACL_SYSTEM_PORT */
 };
 
+static const struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_1b[] = {
+	MLXSW_AFK_ELEMENT_INST_BUF(SRC_IP_0_31, 0x04, 4),
+};
+
 static const struct mlxsw_afk_element_inst mlxsw_sp_afk_element_info_ipv4_5b[] = {
 	MLXSW_AFK_ELEMENT_INST_U32(VIRT_ROUTER, 0x04, 20, 12),
 };

@@ -341,7 +345,7 @@ static const struct mlxsw_afk_block mlxsw_sp4_afk_blocks[] = {
 	MLXSW_AFK_BLOCK(0x14, mlxsw_sp_afk_element_info_mac_4),
 	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x1A, mlxsw_sp_afk_element_info_mac_5b),
 	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x38, mlxsw_sp_afk_element_info_ipv4_0),
-	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x39, mlxsw_sp_afk_element_info_ipv4_1),
+	MLXSW_AFK_BLOCK_HIGH_ENTROPY(0x3F, mlxsw_sp_afk_element_info_ipv4_1b),
 	MLXSW_AFK_BLOCK(0x3A, mlxsw_sp_afk_element_info_ipv4_2),
 	MLXSW_AFK_BLOCK(0x36, mlxsw_sp_afk_element_info_ipv4_5b),
 	MLXSW_AFK_BLOCK(0x40, mlxsw_sp_afk_element_info_ipv6_0),

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #60 lore raw

PATCH 60/356 geneve: do not assume mac header is set in geneve_xmit_skb()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 8588c99c7d47448fcae39e3227d6e2bb97aad86d ]

We should not assume mac header is set in output path.

Use skb_eth_hdr() instead of eth_hdr() to fix the issue.

sysbot reported the following :

 WARNING: CPU: 0 PID: 11635 at include/linux/skbuff.h:3052 skb_mac_header include/linux/skbuff.h:3052 [inline]
 WARNING: CPU: 0 PID: 11635 at include/linux/skbuff.h:3052 eth_hdr include/linux/if_ether.h:24 [inline]
 WARNING: CPU: 0 PID: 11635 at include/linux/skbuff.h:3052 geneve_xmit_skb drivers/net/geneve.c:898 [inline]
 WARNING: CPU: 0 PID: 11635 at include/linux/skbuff.h:3052 geneve_xmit+0x4c38/0x5730 drivers/net/geneve.c:1039
Modules linked in:
CPU: 0 UID: 0 PID: 11635 Comm: syz.4.1423 Not tainted 6.12.0-syzkaller-10296-gaaf20f870da0 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/13/2024
 RIP: 0010:skb_mac_header include/linux/skbuff.h:3052 [inline]
 RIP: 0010:eth_hdr include/linux/if_ether.h:24 [inline]
 RIP: 0010:geneve_xmit_skb drivers/net/geneve.c:898 [inline]
 RIP: 0010:geneve_xmit+0x4c38/0x5730 drivers/net/geneve.c:1039
Code: 21 c6 02 e9 35 d4 ff ff e8 a5 48 4c fb 90 0f 0b 90 e9 fd f5 ff ff e8 97 48 4c fb 90 0f 0b 90 e9 d8 f5 ff ff e8 89 48 4c fb 90 <0f> 0b 90 e9 41 e4 ff ff e8 7b 48 4c fb 90 0f 0b 90 e9 cd e7 ff ff
RSP: 0018:ffffc90003b2f870 EFLAGS: 00010283
RAX: 000000000000037a RBX: 000000000000ffff RCX: ffffc9000dc3d000
RDX: 0000000000080000 RSI: ffffffff86428417 RDI: 0000000000000003
RBP: ffffc90003b2f9f0 R08: 0000000000000003 R09: 000000000000ffff
R10: 000000000000ffff R11: 0000000000000002 R12: ffff88806603c000
R13: 0000000000000000 R14: ffff8880685b2780 R15: 0000000000000e23
FS:  00007fdc2deed6c0(0000) GS:ffff8880b8600000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b30a1dff8 CR3: 0000000056b8c000 CR4: 00000000003526f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
  __netdev_start_xmit include/linux/netdevice.h:5002 [inline]
  netdev_start_xmit include/linux/netdevice.h:5011 [inline]
  __dev_direct_xmit+0x58a/0x720 net/core/dev.c:4490
  dev_direct_xmit include/linux/netdevice.h:3181 [inline]
  packet_xmit+0x1e4/0x360 net/packet/af_packet.c:285
  packet_snd net/packet/af_packet.c:3146 [inline]
  packet_sendmsg+0x2700/0x5660 net/packet/af_packet.c:3178
  sock_sendmsg_nosec net/socket.c:711 [inline]
  __sock_sendmsg net/socket.c:726 [inline]
  __sys_sendto+0x488/0x4f0 net/socket.c:2197
  __do_sys_sendto net/socket.c:2204 [inline]
  __se_sys_sendto net/socket.c:2200 [inline]
  __x64_sys_sendto+0xe0/0x1c0 net/socket.c:2200
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Fixes a025fb5f49ad ("geneve: Allow configuration of DF behaviour")

Reported-by [email protected]

Closes https://lore.kernel.org/netdev/[email protected]/T/#u open here

Reviewed-by Stefano Brivio <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/geneve.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/geneve.c +1 -1

index 8333a5620deff..b939d4711c59b 100644
@@ -986,7 +986,7 @@ static int geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 		if (geneve->cfg.df == GENEVE_DF_SET) {
 			df = htons(IP_DF);
 		} else if (geneve->cfg.df == GENEVE_DF_INHERIT) {
-			struct ethhdr *eth = eth_hdr(skb);
+			struct ethhdr *eth = skb_eth_hdr(skb);
 
 			if (ntohs(eth->h_proto) == ETH_P_IPV6) {
 				df = htons(IP_DF);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #61 lore raw

PATCH 61/356 net/mlx5e: Remove workaround to avoid syndrome for internal port

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jianbo Liu <[email protected]>

[ Upstream commit 5085f861b414e4a51ce28a891dfa32a10a54b64e ]

Previously a workaround was added to avoid syndrome 0xcdb051. It is triggered when offload a rule with tunnel encapsulation, and forwarding to another table, but not matching on the internal port in firmware steering mode. The original workaround skips internal tunnel port logic, which is not correct as not all cases are considered. As an example, if vlan is configured on the uplink port, traffic can't pass because vlan header is not added with this workaround. Besides, there is no such issue for software steering. So, this patch removes that, and returns error directly if trying to offload such rule for firmware steering.

Fixes 06b4eac9c4be ("net/mlx5e: Don't offload internal port if filter device is out device")

Signed-off-by Jianbo Liu <[email protected]>

Tested-by Frode Nordahl <[email protected]>

Reviewed-by Chris Mi <[email protected]>

Reviewed-by Ariel Levkovich <[email protected]>

Signed-off-by Tariq Toukan <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 .../ethernet/mellanox/mlx5/core/en/tc_tun_encap.c   | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c +11 -2

index f1d1e1542e81b..c11092da2e7d8 100644
@@ -5,6 +5,7 @@
 #include <net/nexthop.h>
 #include <net/ip_tunnels.h>
 #include "tc_tun_encap.h"
+#include "fs_core.h"
 #include "en_tc.h"
 #include "tc_tun.h"
 #include "rep/tc.h"

@@ -24,10 +25,18 @@ static int mlx5e_set_int_port_tunnel(struct mlx5e_priv *priv,
 
 	route_dev = dev_get_by_index(dev_net(e->out_dev), e->route_dev_ifindex);
 
-	if (!route_dev || !netif_is_ovs_master(route_dev) ||
-	    attr->parse_attr->filter_dev == e->out_dev)
+	if (!route_dev || !netif_is_ovs_master(route_dev))
 		goto out;
 
+	if (priv->mdev->priv.steering->mode == MLX5_FLOW_STEERING_MODE_DMFS &&
+	    mlx5e_eswitch_uplink_rep(attr->parse_attr->filter_dev) &&
+	    (attr->esw_attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP)) {
+		mlx5_core_warn(priv->mdev,
+			       "Matching on external port with encap + fwd to table actions is not allowed for firmware steering\n");
+		err = -EINVAL;
+		goto out;
+	}
+
 	err = mlx5e_set_fwd_to_int_port_actions(priv, attr, e->route_dev_ifindex,
 						MLX5E_TC_INT_PORT_EGRESS,
 						&attr->action, out_index);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #62 lore raw

PATCH 62/356 net: avoid potential UAF in default_operstate()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 750e51603395e755537da08f745864c93e3ce741 ]

syzbot reported an UAF in default_operstate() [1]

Issue is a race between device and netns dismantles.

After calling __rtnl_unlock() from netdev_run_todo(), we can not assume the netns of each device is still alive.

Make sure the device is not in NETREG_UNREGISTERED state, and add an ASSERT_RTNL() before the call to __dev_get_by_index().

We might move this ASSERT_RTNL() in __dev_get_by_index() in the future.

[1]

BUG: KASAN: slab-use-after-free in __dev_get_by_index+0x5d/0x110 net/core/dev.c:852
Read of size 8 at addr ffff888043eba1b0 by task syz.0.0/5339

CPU: 0 UID: 0 PID: 5339 Comm: syz.0.0 Not tainted 6.12.0-syzkaller-10296-gaaf20f870da0 #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014
Call Trace:
 <TASK>
  __dump_stack lib/dump_stack.c:94 [inline]
  dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120
  print_address_description mm/kasan/report.c:378 [inline]
  print_report+0x169/0x550 mm/kasan/report.c:489
  kasan_report+0x143/0x180 mm/kasan/report.c:602
  __dev_get_by_index+0x5d/0x110 net/core/dev.c:852
  default_operstate net/core/link_watch.c:51 [inline]
  rfc2863_policy+0x224/0x300 net/core/link_watch.c:67
  linkwatch_do_dev+0x3e/0x170 net/core/link_watch.c:170
  netdev_run_todo+0x461/0x1000 net/core/dev.c:10894
  rtnl_unlock net/core/rtnetlink.c:152 [inline]
  rtnl_net_unlock include/linux/rtnetlink.h:133 [inline]
  rtnl_dellink+0x760/0x8d0 net/core/rtnetlink.c:3520
  rtnetlink_rcv_msg+0x791/0xcf0 net/core/rtnetlink.c:6911
  netlink_rcv_skb+0x1e3/0x430 net/netlink/af_netlink.c:2541
  netlink_unicast_kernel net/netlink/af_netlink.c:1321 [inline]
  netlink_unicast+0x7f6/0x990 net/netlink/af_netlink.c:1347
  netlink_sendmsg+0x8e4/0xcb0 net/netlink/af_netlink.c:1891
  sock_sendmsg_nosec net/socket.c:711 [inline]
  __sock_sendmsg+0x221/0x270 net/socket.c:726
  ____sys_sendmsg+0x52a/0x7e0 net/socket.c:2583
  ___sys_sendmsg net/socket.c:2637 [inline]
  __sys_sendmsg+0x269/0x350 net/socket.c:2669
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f2a3cb80809
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f2a3d9cd058 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: ffffffffffffffda RBX: 00007f2a3cd45fa0 RCX: 00007f2a3cb80809
RDX: 0000000000000000 RSI: 0000000020000000 RDI: 0000000000000008
RBP: 00007f2a3cbf393e R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 0000000000000000 R14: 00007f2a3cd45fa0 R15: 00007ffd03bc65c8
 </TASK>

Allocated by task 5339:
  kasan_save_stack mm/kasan/common.c:47 [inline]
  kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
  poison_kmalloc_redzone mm/kasan/common.c:377 [inline]
  __kasan_kmalloc+0x98/0xb0 mm/kasan/common.c:394
  kasan_kmalloc include/linux/kasan.h:260 [inline]
  __kmalloc_cache_noprof+0x243/0x390 mm/slub.c:4314
  kmalloc_noprof include/linux/slab.h:901 [inline]
  kmalloc_array_noprof include/linux/slab.h:945 [inline]
  netdev_create_hash net/core/dev.c:11870 [inline]
  netdev_init+0x10c/0x250 net/core/dev.c:11890
  ops_init+0x31e/0x590 net/core/net_namespace.c:138
  setup_net+0x287/0x9e0 net/core/net_namespace.c:362
  copy_net_ns+0x33f/0x570 net/core/net_namespace.c:500
  create_new_namespaces+0x425/0x7b0 kernel/nsproxy.c:110
  unshare_nsproxy_namespaces+0x124/0x180 kernel/nsproxy.c:228
  ksys_unshare+0x57d/0xa70 kernel/fork.c:3314
  __do_sys_unshare kernel/fork.c:3385 [inline]
  __se_sys_unshare kernel/fork.c:3383 [inline]
  __x64_sys_unshare+0x38/0x40 kernel/fork.c:3383
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 12:
  kasan_save_stack mm/kasan/common.c:47 [inline]
  kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
  kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:582
  poison_slab_object mm/kasan/common.c:247 [inline]
  __kasan_slab_free+0x59/0x70 mm/kasan/common.c:264
  kasan_slab_free include/linux/kasan.h:233 [inline]
  slab_free_hook mm/slub.c:2338 [inline]
  slab_free mm/slub.c:4598 [inline]
  kfree+0x196/0x420 mm/slub.c:4746
  netdev_exit+0x65/0xd0 net/core/dev.c:11992
  ops_exit_list net/core/net_namespace.c:172 [inline]
  cleanup_net+0x802/0xcc0 net/core/net_namespace.c:632
  process_one_work kernel/workqueue.c:3229 [inline]
  process_scheduled_works+0xa63/0x1850 kernel/workqueue.c:3310
  worker_thread+0x870/0xd30 kernel/workqueue.c:3391
  kthread+0x2f0/0x390 kernel/kthread.c:389
  ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
  ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244

The buggy address belongs to the object at ffff888043eba000
 which belongs to the cache kmalloc-2k of size 2048

The buggy address is located 432 bytes inside of
freed 2048-byte region [ffff888043eba000, ffff888043eba800)

The buggy address belongs to the physical page:
page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x43eb8
head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0x4fff00000000040(head|node=1|zone=1|lastcpupid=0x7ff)
page_type: f5(slab)
raw: 04fff00000000040 ffff88801ac42000 dead000000000122 0000000000000000
raw: 0000000000000000 0000000000080008 00000001f5000000 0000000000000000

head: 04fff00000000040 ffff88801ac42000 dead000000000122 0000000000000000
head: 0000000000000000 0000000000080008 00000001f5000000 0000000000000000
head: 04fff00000000003 ffffea00010fae01 ffffffffffffffff 0000000000000000
head: 0000000000000008 0000000000000000 00000000ffffffff 0000000000000000

page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 5339, tgid 5338 (syz.0.0), ts 69674195892, free_ts 69663220888
  set_page_owner include/linux/page_owner.h:32 [inline]
  post_alloc_hook+0x1f3/0x230 mm/page_alloc.c:1556
  prep_new_page mm/page_alloc.c:1564 [inline]
  get_page_from_freelist+0x3649/0x3790 mm/page_alloc.c:3474
  __alloc_pages_noprof+0x292/0x710 mm/page_alloc.c:4751
  alloc_pages_mpol_noprof+0x3e8/0x680 mm/mempolicy.c:2265
  alloc_slab_page+0x6a/0x140 mm/slub.c:2408
  allocate_slab+0x5a/0x2f0 mm/slub.c:2574
  new_slab mm/slub.c:2627 [inline]
  ___slab_alloc+0xcd1/0x14b0 mm/slub.c:3815
  __slab_alloc+0x58/0xa0 mm/slub.c:3905
  __slab_alloc_node mm/slub.c:3980 [inline]
  slab_alloc_node mm/slub.c:4141 [inline]
  __do_kmalloc_node mm/slub.c:4282 [inline]
  __kmalloc_noprof+0x2e6/0x4c0 mm/slub.c:4295
  kmalloc_noprof include/linux/slab.h:905 [inline]
  sk_prot_alloc+0xe0/0x210 net/core/sock.c:2165
  sk_alloc+0x38/0x370 net/core/sock.c:2218
  __netlink_create+0x65/0x260 net/netlink/af_netlink.c:629
  __netlink_kernel_create+0x174/0x6f0 net/netlink/af_netlink.c:2015
  netlink_kernel_create include/linux/netlink.h:62 [inline]
  uevent_net_init+0xed/0x2d0 lib/kobject_uevent.c:783
  ops_init+0x31e/0x590 net/core/net_namespace.c:138
  setup_net+0x287/0x9e0 net/core/net_namespace.c:362
page last free pid 1032 tgid 1032 stack trace:
  reset_page_owner include/linux/page_owner.h:25 [inline]
  free_pages_prepare mm/page_alloc.c:1127 [inline]
  free_unref_page+0xdf9/0x1140 mm/page_alloc.c:2657
  __slab_free+0x31b/0x3d0 mm/slub.c:4509
  qlink_free mm/kasan/quarantine.c:163 [inline]
  qlist_free_all+0x9a/0x140 mm/kasan/quarantine.c:179
  kasan_quarantine_reduce+0x14f/0x170 mm/kasan/quarantine.c:286
  __kasan_slab_alloc+0x23/0x80 mm/kasan/common.c:329
  kasan_slab_alloc include/linux/kasan.h:250 [inline]
  slab_post_alloc_hook mm/slub.c:4104 [inline]
  slab_alloc_node mm/slub.c:4153 [inline]
  kmem_cache_alloc_node_noprof+0x1d9/0x380 mm/slub.c:4205
  __alloc_skb+0x1c3/0x440 net/core/skbuff.c:668
  alloc_skb include/linux/skbuff.h:1323 [inline]
  alloc_skb_with_frags+0xc3/0x820 net/core/skbuff.c:6612
  sock_alloc_send_pskb+0x91a/0xa60 net/core/sock.c:2881
  sock_alloc_send_skb include/net/sock.h:1797 [inline]
  mld_newpack+0x1c3/0xaf0 net/ipv6/mcast.c:1747
  add_grhead net/ipv6/mcast.c:1850 [inline]
  add_grec+0x1492/0x19a0 net/ipv6/mcast.c:1988
  mld_send_initial_cr+0x228/0x4b0 net/ipv6/mcast.c:2234
  ipv6_mc_dad_complete+0x88/0x490 net/ipv6/mcast.c:2245
  addrconf_dad_completed+0x712/0xcd0 net/ipv6/addrconf.c:4342
 addrconf_dad_work+0xdc2/0x16f0
  process_one_work kernel/workqueue.c:3229 [inline]
  process_scheduled_works+0xa63/0x1850 kernel/workqueue.c:3310

Memory state around the buggy address:
 ffff888043eba080: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff888043eba100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff888043eba180: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                     ^
 ffff888043eba200: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff888043eba280: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb

Fixes 8c55facecd7a ("net: linkwatch: only report IF_OPER_LOWERLAYERDOWN if iflink is actually down")

Reported-by [email protected]

Closes https://lore.kernel.org/netdev/[email protected]/T/#u open here

Reviewed-by Vladimir Oltean <[email protected]>

Link https://patch.msgid.link/[email protected] open here

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/core/link_watch.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

net/core/link_watch.c +6 -1

index cf867f6e38bf1..66422c95c83c7 100644
@@ -45,9 +45,14 @@ static unsigned char default_operstate(const struct net_device *dev)
 		int iflink = dev_get_iflink(dev);
 		struct net_device *peer;
 
-		if (iflink == dev->ifindex)
+		/* If called from netdev_run_todo()/linkwatch_sync_dev(),
+		 * dev_net(dev) can be already freed, and RTNL is not held.
+		 */
+		if (dev->reg_state == NETREG_UNREGISTERED ||
+		    iflink == dev->ifindex)
 			return IF_OPER_DOWN;
 
+		ASSERT_RTNL();
 		peer = __dev_get_by_index(dev_net(dev), iflink);
 		if (!peer)
 			return IF_OPER_DOWN;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #63 lore raw

PATCH 63/356 KVM: arm64: Change kvm_handle_mmio_return() return polarity

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Fuad Tabba <[email protected]>

[ Upstream commit cc81b6dfc3bc82c3a2600eefbd3823bdb2190197 ]

Most exit handlers return <= 0 to indicate that the host needs to handle the exit. Make kvm_handle_mmio_return() consistent with the exit handlers in handle_exit(). This makes the code easier to reason about, and makes it easier to add other handlers in future patches.

No functional change intended.

Signed-off-by Fuad Tabba <[email protected]>

Acked-by Oliver Upton <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Marc Zyngier <[email protected]>

Stable-dep-of: e735a5da6442 ("KVM: arm64: Don't retire aborted MMIO instruction")

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/kvm/arm.c  | 2 +-
 arch/arm64/kvm/mmio.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

arch/arm64/kvm/arm.c +1 -1

index 4742e6c5ea7a0..ffdc2c4d07ee8 100644
@@ -900,7 +900,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu)
 
 	if (run->exit_reason == KVM_EXIT_MMIO) {
 		ret = kvm_handle_mmio_return(vcpu);
-		if (ret)
+		if (ret <= 0)
 			return ret;
 	}

arch/arm64/kvm/mmio.c +2 -2

index 3dd38a151d2a6..886ef30e12196 100644
@@ -86,7 +86,7 @@ int kvm_handle_mmio_return(struct kvm_vcpu *vcpu)
 
 	/* Detect an already handled MMIO return */
 	if (unlikely(!vcpu->mmio_needed))
-		return 0;
+		return 1;
 
 	vcpu->mmio_needed = 0;

@@ -117,7 +117,7 @@ int kvm_handle_mmio_return(struct kvm_vcpu *vcpu)
 	 */
 	kvm_incr_pc(vcpu);
 
-	return 0;
+	return 1;
 }
 
 int io_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #64 lore raw

PATCH 64/356 KVM: arm64: Dont retire aborted MMIO instruction

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Oliver Upton <[email protected]>

[ Upstream commit e735a5da64420a86be370b216c269b5dd8e830e2 ]

Returning an abort to the guest for an unsupported MMIO access is a documented feature of the KVM UAPI. Nevertheless, it's clear that this plumbing has seen limited testing, since userspace can trivially cause a WARN in the MMIO return:

  WARNING: CPU: 0 PID: 30558 at arch/arm64/include/asm/kvm_emulate.h:536 kvm_handle_mmio_return+0x46c/0x5c4 arch/arm64/include/asm/kvm_emulate.h:536
  Call trace:
   kvm_handle_mmio_return+0x46c/0x5c4 arch/arm64/include/asm/kvm_emulate.h:536
   kvm_arch_vcpu_ioctl_run+0x98/0x15b4 arch/arm64/kvm/arm.c:1133
   kvm_vcpu_ioctl+0x75c/0xa78 virt/kvm/kvm_main.c:4487
   __do_sys_ioctl fs/ioctl.c:51 [inline]
   __se_sys_ioctl fs/ioctl.c:893 [inline]
   __arm64_sys_ioctl+0x14c/0x1c8 fs/ioctl.c:893
   __invoke_syscall arch/arm64/kernel/syscall.c:35 [inline]
   invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:49
   el0_svc_common+0x1e0/0x23c arch/arm64/kernel/syscall.c:132
   do_el0_svc+0x48/0x58 arch/arm64/kernel/syscall.c:151
   el0_svc+0x38/0x68 arch/arm64/kernel/entry-common.c:712
   el0t_64_sync_handler+0x90/0xfc arch/arm64/kernel/entry-common.c:730
   el0t_64_sync+0x190/0x194 arch/arm64/kernel/entry.S:598

The splat is complaining that KVM is advancing PC while an exception is pending, i.e. that KVM is retiring the MMIO instruction despite a pending synchronous external abort. Womp womp.

Fix the glaring UAPI bug by skipping over all the MMIO emulation in case there is a pending synchronous exception. Note that while userspace is capable of pending an asynchronous exception (SError, IRQ, or FIQ), it is still safe to retire the MMIO instruction in this case as (1) they are by definition asynchronous, and (2) KVM relies on hardware support for pending/delivering these exceptions instead of the software state machine for advancing PC.

Fixes da345174ceca ("KVM: arm/arm64: Allow user injection of external data aborts")

Reported-by Alexander Potapenko <[email protected]>

Reviewed-by Marc Zyngier <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Oliver Upton <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/kvm/mmio.c | 32 ++++++++++++++++++++++++++++++--
 1 file changed, 30 insertions(+), 2 deletions(-)

arch/arm64/kvm/mmio.c +30 -2

index 886ef30e12196..2aa503ff742ee 100644
@@ -72,6 +72,31 @@ unsigned long kvm_mmio_read_buf(const void *buf, unsigned int len)
 	return data;
 }
 
+static bool kvm_pending_sync_exception(struct kvm_vcpu *vcpu)
+{
+	if (!vcpu_get_flag(vcpu, PENDING_EXCEPTION))
+		return false;
+
+	if (vcpu_el1_is_32bit(vcpu)) {
+		switch (vcpu_get_flag(vcpu, EXCEPT_MASK)) {
+		case unpack_vcpu_flag(EXCEPT_AA32_UND):
+		case unpack_vcpu_flag(EXCEPT_AA32_IABT):
+		case unpack_vcpu_flag(EXCEPT_AA32_DABT):
+			return true;
+		default:
+			return false;
+		}
+	} else {
+		switch (vcpu_get_flag(vcpu, EXCEPT_MASK)) {
+		case unpack_vcpu_flag(EXCEPT_AA64_EL1_SYNC):
+		case unpack_vcpu_flag(EXCEPT_AA64_EL2_SYNC):
+			return true;
+		default:
+			return false;
+		}
+	}
+}
+
 /**
  * kvm_handle_mmio_return -- Handle MMIO loads after user space emulation
  *			     or in-kernel IO emulation

@@ -84,8 +109,11 @@ int kvm_handle_mmio_return(struct kvm_vcpu *vcpu)
 	unsigned int len;
 	int mask;
 
-	/* Detect an already handled MMIO return */
-	if (unlikely(!vcpu->mmio_needed))
+	/*
+	 * Detect if the MMIO return was already handled or if userspace aborted
+	 * the MMIO access.
+	 */
+	if (unlikely(!vcpu->mmio_needed || kvm_pending_sync_exception(vcpu)))
 		return 1;
 
 	vcpu->mmio_needed = 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #65 lore raw

PATCH 65/356 xhci: Allow RPM on the USB controller (1022:43f7) by default

Link https://lore.kernel.org/all/[email protected]/ open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Basavaraj Natikar <[email protected]>

[ Upstream commit 28cbed496059fe1868203b76e9e0ef285733524d ]

Enable runtime PM by default for older AMD 1022:43f7 xHCI 1.1 host as it is proven to work.
Driver enables runtime PM by default for newer xHCI 1.2 host.

Cc Mario Limonciello <[email protected]>

Tested-by Oleksandr Natalenko <[email protected]>

Signed-off-by Basavaraj Natikar <[email protected]>

Acked-by Mathias Nyman <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Stable-dep-of: d7b11fe57902 ("xhci: Combine two if statements for Etron xHCI host")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/usb/host/xhci-pci.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

drivers/usb/host/xhci-pci.c +4 -1

index c2b37aa2cdfca..3a2a0d8f7af86 100644
@@ -374,8 +374,11 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 		xhci->quirks |= XHCI_RESET_ON_RESUME;
 	}
 
-	if (pdev->vendor == PCI_VENDOR_ID_AMD)
+	if (pdev->vendor == PCI_VENDOR_ID_AMD) {
 		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
+		if (pdev->device == 0x43f7)
+			xhci->quirks |= XHCI_DEFAULT_PM_RUNTIME_ALLOW;
+	}
 
 	if ((pdev->vendor == PCI_VENDOR_ID_AMD) &&
 		((pdev->device == PCI_DEVICE_ID_AMD_PROMONTORYA_4) ||

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #66 lore raw

PATCH 66/356 xhci: remove XHCI_TRUST_TX_LENGTH quirk

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mathias Nyman <[email protected]>

[ Upstream commit 34b67198244f2d7d8409fa4eb76204c409c0c97e ]

If this quirk was set then driver would treat transfer events with 'Success' completion code as 'Short packet' if there were untransferred bytes left.

This is so common that turn it into default behavior.

xhci_warn_ratelimited() is no longer used after this, so remove it.

A success event with untransferred bytes left doesn't always mean a misbehaving controller. If there was an error mid a multi-TRB TD it's allowed to issue a success event for the last TRB in that TD.

See xhci 1.2 spec 4.9.1 Transfer Descriptors

"Note: If an error is detected while processing a multi-TRB TD, the xHC
shall generate a Transfer Event for the TRB that the error was detected on with the appropriate error Condition Code, then may advance to the next TD. If in the process of advancing to the next TD, a Transfer TRB is encountered with its IOC flag set, then the Condition Code of the Transfer Event generated for that Transfer TRB should be Success, because there was no error actually associated with the TRB that generated the Event. However, an xHC implementation may redundantly assert the original error Condition Code."

Co-developed-by Niklas Neronin <[email protected]>

Signed-off-by Niklas Neronin <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Stable-dep-of: d7b11fe57902 ("xhci: Combine two if statements for Etron xHCI host")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/usb/host/xhci-pci.c  | 15 ++-------------
 drivers/usb/host/xhci-rcar.c |  6 ++----
 drivers/usb/host/xhci-ring.c | 15 +++++----------
 drivers/usb/host/xhci.h      |  4 +---
 4 files changed, 10 insertions(+), 30 deletions(-)

drivers/usb/host/xhci-pci.c +2 ~~-13~~

index 3a2a0d8f7af86..f74c067b54587 100644
@@ -337,17 +337,12 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 				"QUIRK: Fresco Logic revision %u "
 				"has broken MSI implementation",
 				pdev->revision);
-		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
 	}
 
 	if (pdev->vendor == PCI_VENDOR_ID_FRESCO_LOGIC &&
 			pdev->device == PCI_DEVICE_ID_FRESCO_LOGIC_FL1009)
 		xhci->quirks |= XHCI_BROKEN_STREAMS;
 
-	if (pdev->vendor == PCI_VENDOR_ID_FRESCO_LOGIC &&
-			pdev->device == PCI_DEVICE_ID_FRESCO_LOGIC_FL1100)
-		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
-
 	if (pdev->vendor == PCI_VENDOR_ID_NEC)
 		xhci->quirks |= XHCI_NEC_HOST;

@@ -374,11 +369,8 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 		xhci->quirks |= XHCI_RESET_ON_RESUME;
 	}
 
-	if (pdev->vendor == PCI_VENDOR_ID_AMD) {
-		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
-		if (pdev->device == 0x43f7)
-			xhci->quirks |= XHCI_DEFAULT_PM_RUNTIME_ALLOW;
-	}
+	if (pdev->vendor == PCI_VENDOR_ID_AMD && pdev->device == 0x43f7)
+		xhci->quirks |= XHCI_DEFAULT_PM_RUNTIME_ALLOW;
 
 	if ((pdev->vendor == PCI_VENDOR_ID_AMD) &&
 		((pdev->device == PCI_DEVICE_ID_AMD_PROMONTORYA_4) ||

@@ -466,7 +458,6 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 	if (pdev->vendor == PCI_VENDOR_ID_ETRON &&
 			pdev->device == PCI_DEVICE_ID_EJ168) {
 		xhci->quirks |= XHCI_RESET_ON_RESUME;
-		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
 		xhci->quirks |= XHCI_BROKEN_STREAMS;
 	}
 	if (pdev->vendor == PCI_VENDOR_ID_ETRON &&

@@ -477,7 +468,6 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 
 	if (pdev->vendor == PCI_VENDOR_ID_RENESAS &&
 	    pdev->device == 0x0014) {
-		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
 		xhci->quirks |= XHCI_ZERO_64B_REGS;
 	}
 	if (pdev->vendor == PCI_VENDOR_ID_RENESAS &&

@@ -507,7 +497,6 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 	}
 	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA &&
 		pdev->device == PCI_DEVICE_ID_ASMEDIA_1042A_XHCI) {
-		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
 		xhci->quirks |= XHCI_NO_64BIT_SUPPORT;
 	}
 	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA &&

drivers/usb/host/xhci-rcar.c +2 -4

index ab9c5969e4624..8b357647728c2 100644
@@ -214,8 +214,7 @@ static int xhci_rcar_resume_quirk(struct usb_hcd *hcd)
  */
 #define SET_XHCI_PLAT_PRIV_FOR_RCAR(firmware)				\
 	.firmware_name = firmware,					\
-	.quirks = XHCI_NO_64BIT_SUPPORT | XHCI_TRUST_TX_LENGTH |	\
-		  XHCI_SLOW_SUSPEND,					\
+	.quirks = XHCI_NO_64BIT_SUPPORT |  XHCI_SLOW_SUSPEND,		\
 	.init_quirk = xhci_rcar_init_quirk,				\
 	.plat_start = xhci_rcar_start,					\
 	.resume_quirk = xhci_rcar_resume_quirk,

@@ -229,8 +228,7 @@ static const struct xhci_plat_priv xhci_plat_renesas_rcar_gen3 = {
 };
 
 static const struct xhci_plat_priv xhci_plat_renesas_rzv2m = {
-	.quirks = XHCI_NO_64BIT_SUPPORT | XHCI_TRUST_TX_LENGTH |
-		  XHCI_SLOW_SUSPEND,
+	.quirks = XHCI_NO_64BIT_SUPPORT | XHCI_SLOW_SUSPEND,
 	.init_quirk = xhci_rzv2m_init_quirk,
 	.plat_start = xhci_rzv2m_start,
 };

drivers/usb/host/xhci-ring.c +5 ~~-10~~

index 258e64d6522c6..99759926daac6 100644
@@ -2443,8 +2443,7 @@ static int process_isoc_td(struct xhci_hcd *xhci, struct xhci_virt_ep *ep,
 			break;
 		if (remaining) {
 			frame->status = short_framestatus;
-			if (xhci->quirks & XHCI_TRUST_TX_LENGTH)
-				sum_trbs_for_length = true;
+			sum_trbs_for_length = true;
 			break;
 		}
 		frame->status = 0;

@@ -2693,15 +2692,11 @@ static int handle_tx_event(struct xhci_hcd *xhci,
 	 * transfer type
 	 */
 	case COMP_SUCCESS:
-		if (EVENT_TRB_LEN(le32_to_cpu(event->transfer_len)) == 0)
-			break;
-		if (xhci->quirks & XHCI_TRUST_TX_LENGTH ||
-		    ep_ring->last_td_was_short)
+		if (EVENT_TRB_LEN(le32_to_cpu(event->transfer_len)) != 0) {
 			trb_comp_code = COMP_SHORT_PACKET;
-		else
-			xhci_warn_ratelimited(xhci,
-					      "WARN Successful completion on short TX for slot %u ep %u: needs XHCI_TRUST_TX_LENGTH quirk?\n",
-					      slot_id, ep_index);
+			xhci_dbg(xhci, "Successful completion on short TX for slot %u ep %u with last td short %d\n",
+				 slot_id, ep_index, ep_ring->last_td_was_short);
+		}
 		break;
 	case COMP_SHORT_PACKET:
 		break;

drivers/usb/host/xhci.h +1 -3

index f2190d121233b..a0005a1124938 100644
@@ -1619,7 +1619,7 @@ struct xhci_hcd {
 #define XHCI_RESET_ON_RESUME	BIT_ULL(7)
 #define	XHCI_SW_BW_CHECKING	BIT_ULL(8)
 #define XHCI_AMD_0x96_HOST	BIT_ULL(9)
-#define XHCI_TRUST_TX_LENGTH	BIT_ULL(10)
+#define XHCI_TRUST_TX_LENGTH	BIT_ULL(10) /* Deprecated */
 #define XHCI_LPM_SUPPORT	BIT_ULL(11)
 #define XHCI_INTEL_HOST		BIT_ULL(12)
 #define XHCI_SPURIOUS_REBOOT	BIT_ULL(13)

@@ -1761,8 +1761,6 @@ static inline bool xhci_has_one_roothub(struct xhci_hcd *xhci)
 	dev_err(xhci_to_hcd(xhci)->self.controller , fmt , ## args)
 #define xhci_warn(xhci, fmt, args...) \
 	dev_warn(xhci_to_hcd(xhci)->self.controller , fmt , ## args)
-#define xhci_warn_ratelimited(xhci, fmt, args...) \
-	dev_warn_ratelimited(xhci_to_hcd(xhci)->self.controller , fmt , ## args)
 #define xhci_info(xhci, fmt, args...) \
 	dev_info(xhci_to_hcd(xhci)->self.controller , fmt , ## args)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #67 lore raw

PATCH 67/356 xhci: Combine two if statements for Etron xHCI host

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuangyi Chiang <[email protected]>

[ Upstream commit d7b11fe5790203fcc0db182249d7bfd945e44ccb ]

Combine two if statements, because these hosts have the same quirk flags applied.

[Mathias: has stable tag because other fixes in series depend on this]

Fixes 91f7a1524a92 ("xhci: Apply broken streams quirk to Etron EJ188 xHCI host")

Signed-off-by Kuangyi Chiang <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/usb/host/xhci-pci.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

drivers/usb/host/xhci-pci.c +2 -6

index f74c067b54587..d36158df83afc 100644
@@ -456,12 +456,8 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 		xhci->quirks |= XHCI_DEFAULT_PM_RUNTIME_ALLOW;
 
 	if (pdev->vendor == PCI_VENDOR_ID_ETRON &&
-			pdev->device == PCI_DEVICE_ID_EJ168) {
-		xhci->quirks |= XHCI_RESET_ON_RESUME;
-		xhci->quirks |= XHCI_BROKEN_STREAMS;
-	}
-	if (pdev->vendor == PCI_VENDOR_ID_ETRON &&
-			pdev->device == PCI_DEVICE_ID_EJ188) {
+	    (pdev->device == PCI_DEVICE_ID_EJ168 ||
+	     pdev->device == PCI_DEVICE_ID_EJ188)) {
 		xhci->quirks |= XHCI_RESET_ON_RESUME;
 		xhci->quirks |= XHCI_BROKEN_STREAMS;
 	}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #68 lore raw

PATCH 68/356 xhci: Dont issue Reset Device command to Etron xHCI host

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuangyi Chiang <[email protected]>

[ Upstream commit 76d98856b1c6d06ce18f32c20527a4f9d283e660 ]

Sometimes the hub driver does not recognize the USB device connected to the external USB2.0 hub when the system resumes from S4.

After the SetPortFeature(PORT_RESET) request is completed, the hub driver calls the HCD reset_device callback, which will issue a Reset Device command and free all structures associated with endpoints that were disabled.

This happens when the xHCI driver issue a Reset Device command to inform the Etron xHCI host that the USB device associated with a device slot has been reset. Seems that the Etron xHCI host can not perform this command correctly, affecting the USB device.

To work around this, the xHCI driver should obtain a new device slot with reference to commit 651aaf36a7d7 ("usb: xhci: Handle USB transaction error on address command"), which is another way to inform the Etron xHCI host that the USB device has been reset.

Add a new XHCI_ETRON_HOST quirk flag to invoke the workaround in xhci_discover_or_reset_device().

Fixes 2a8f82c4ceaf ("USB: xhci: Notify the xHC when a device is reset.")

Signed-off-by Kuangyi Chiang <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/usb/host/xhci-pci.c |  1 +
 drivers/usb/host/xhci.c     | 19 +++++++++++++++++++
 drivers/usb/host/xhci.h     |  1 +
 3 files changed, 21 insertions(+)

drivers/usb/host/xhci-pci.c +1 -0

index d36158df83afc..340d9597d1ab0 100644
@@ -458,6 +458,7 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
 	if (pdev->vendor == PCI_VENDOR_ID_ETRON &&
 	    (pdev->device == PCI_DEVICE_ID_EJ168 ||
 	     pdev->device == PCI_DEVICE_ID_EJ188)) {
+		xhci->quirks |= XHCI_ETRON_HOST;
 		xhci->quirks |= XHCI_RESET_ON_RESUME;
 		xhci->quirks |= XHCI_BROKEN_STREAMS;
 	}

drivers/usb/host/xhci.c +19 -0

index f005ce1f91ca2..3bd70e6ad64ba 100644
@@ -3642,6 +3642,8 @@ void xhci_free_device_endpoint_resources(struct xhci_hcd *xhci,
 				xhci->num_active_eps);
 }
 
+static void xhci_free_dev(struct usb_hcd *hcd, struct usb_device *udev);
+
 /*
  * This submits a Reset Device Command, which will set the device state to 0,
  * set the device address to 0, and disable all the endpoints except the default

@@ -3712,6 +3714,23 @@ static int xhci_discover_or_reset_device(struct usb_hcd *hcd,
 						SLOT_STATE_DISABLED)
 		return 0;
 
+	if (xhci->quirks & XHCI_ETRON_HOST) {
+		/*
+		 * Obtaining a new device slot to inform the xHCI host that
+		 * the USB device has been reset.
+		 */
+		ret = xhci_disable_slot(xhci, udev->slot_id);
+		xhci_free_virt_device(xhci, udev->slot_id);
+		if (!ret) {
+			ret = xhci_alloc_dev(hcd, udev);
+			if (ret == 1)
+				ret = 0;
+			else
+				ret = -EINVAL;
+		}
+		return ret;
+	}
+
 	trace_xhci_discover_or_reset_device(slot_ctx);
 
 	xhci_dbg(xhci, "Resetting device with slot ID %u\n", slot_id);

drivers/usb/host/xhci.h +1 -0

index a0005a1124938..4bbd12db7239a 100644
@@ -1660,6 +1660,7 @@ struct xhci_hcd {
 #define XHCI_ZHAOXIN_HOST	BIT_ULL(46)
 #define XHCI_WRITE_64_HI_LO	BIT_ULL(47)
 #define XHCI_CDNS_SCTX_QUIRK	BIT_ULL(48)
+#define XHCI_ETRON_HOST	BIT_ULL(49)
 
 	unsigned int		num_active_eps;
 	unsigned int		limit_active_eps;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #69 lore raw

PATCH 69/356 xhci: Fix control transfer error on Etron xHCI host

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuangyi Chiang <[email protected]>

[ Upstream commit 5e1c67abc9301d05130b7e267c204e7005503b33 ]

Performing a stability stress test on a USB3.0 2.5G ethernet adapter results in errors like this:

[   91.441469] r8152 2-3:1.0 eth3: get_registers -71
[   91.458659] r8152 2-3:1.0 eth3: get_registers -71
[   91.475911] r8152 2-3:1.0 eth3: get_registers -71
[   91.493203] r8152 2-3:1.0 eth3: get_registers -71
[   91.510421] r8152 2-3:1.0 eth3: get_registers -71

The r8152 driver will periodically issue lots of control-IN requests to access the status of ethernet adapter hardware registers during the test.

This happens when the xHCI driver enqueue a control TD (which cross over the Link TRB between two ring segments, as shown) in the endpoint zero's transfer ring. Seems the Etron xHCI host can not perform this TD correctly, causing the USB transfer error occurred, maybe the upper driver retry that control-IN request can solve problem, but not all drivers do this.

| |
-------
| TRB | Setup Stage
-------
| TRB | Link
-------
-------
| TRB | Data Stage
-------
| TRB | Status Stage
-------
| |

To work around this, the xHCI driver should enqueue a No Op TRB if next available TRB is the Link TRB in the ring segment, this can prevent the Setup and Data Stage TRB to be breaked by the Link TRB.

Check if the XHCI_ETRON_HOST quirk flag is set before invoking the workaround in xhci_queue_ctrl_tx().

Fixes d0e96f5a71a0 ("USB: xhci: Control transfer support.")

Signed-off-by Kuangyi Chiang <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/usb/host/xhci-ring.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

drivers/usb/host/xhci-ring.c +14 -0

index 99759926daac6..50f5880114004 100644
@@ -3828,6 +3828,20 @@ int xhci_queue_ctrl_tx(struct xhci_hcd *xhci, gfp_t mem_flags,
 	if (!urb->setup_packet)
 		return -EINVAL;
 
+	if ((xhci->quirks & XHCI_ETRON_HOST) &&
+	    urb->dev->speed >= USB_SPEED_SUPER) {
+		/*
+		 * If next available TRB is the Link TRB in the ring segment then
+		 * enqueue a No Op TRB, this can prevent the Setup and Data Stage
+		 * TRB to be breaked by the Link TRB.
+		 */
+		if (trb_is_link(ep_ring->enqueue + 1)) {
+			field = TRB_TYPE(TRB_TR_NOOP) | ep_ring->cycle_state;
+			queue_trb(xhci, ep_ring, false, 0, 0,
+					TRB_INTR_TARGET(0), field);
+		}
+	}
+
 	/* 1 TRB for setup, 1 for status */
 	num_trbs = 2;
 	/*

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:33 UTC #70 lore raw

PATCH 70/356 gpio: grgpio: use a helper variable to store the address of ofdev->dev

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bartosz Golaszewski <[email protected]>

[ Upstream commit d036ae41cebdfae92666024163c109b8fef516fa ]

Instead of dereferencing the platform device pointer repeatedly, just store its address in a helper variable.

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Bartosz Golaszewski <[email protected]>

Stable-dep-of: 050b23d081da ("gpio: grgpio: Add NULL check in grgpio_probe")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpio/gpio-grgpio.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

drivers/gpio/gpio-grgpio.c +12 ~~-11~~

index 0163c95f6dd75..fe919d9bd46a3 100644
@@ -328,6 +328,7 @@ static const struct irq_domain_ops grgpio_irq_domain_ops = {
 static int grgpio_probe(struct platform_device *ofdev)
 {
 	struct device_node *np = ofdev->dev.of_node;
+	struct device *dev = &ofdev->dev;
 	void  __iomem *regs;
 	struct gpio_chip *gc;
 	struct grgpio_priv *priv;

@@ -337,7 +338,7 @@ static int grgpio_probe(struct platform_device *ofdev)
 	int size;
 	int i;
 
-	priv = devm_kzalloc(&ofdev->dev, sizeof(*priv), GFP_KERNEL);
+	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
 	if (!priv)
 		return -ENOMEM;

@@ -346,28 +347,28 @@ static int grgpio_probe(struct platform_device *ofdev)
 		return PTR_ERR(regs);
 
 	gc = &priv->gc;
-	err = bgpio_init(gc, &ofdev->dev, 4, regs + GRGPIO_DATA,
+	err = bgpio_init(gc, dev, 4, regs + GRGPIO_DATA,
 			 regs + GRGPIO_OUTPUT, NULL, regs + GRGPIO_DIR, NULL,
 			 BGPIOF_BIG_ENDIAN_BYTE_ORDER);
 	if (err) {
-		dev_err(&ofdev->dev, "bgpio_init() failed\n");
+		dev_err(dev, "bgpio_init() failed\n");
 		return err;
 	}
 
 	priv->regs = regs;
 	priv->imask = gc->read_reg(regs + GRGPIO_IMASK);
-	priv->dev = &ofdev->dev;
+	priv->dev = dev;
 
 	gc->owner = THIS_MODULE;
 	gc->to_irq = grgpio_to_irq;
-	gc->label = devm_kasprintf(&ofdev->dev, GFP_KERNEL, "%pOF", np);
+	gc->label = devm_kasprintf(dev, GFP_KERNEL, "%pOF", np);
 	gc->base = -1;
 
 	err = of_property_read_u32(np, "nbits", &prop);
 	if (err || prop <= 0 || prop > GRGPIO_MAX_NGPIO) {
 		gc->ngpio = GRGPIO_MAX_NGPIO;
-		dev_dbg(&ofdev->dev,
-			"No or invalid nbits property: assume %d\n", gc->ngpio);
+		dev_dbg(dev, "No or invalid nbits property: assume %d\n",
+			gc->ngpio);
 	} else {
 		gc->ngpio = prop;
 	}

@@ -379,7 +380,7 @@ static int grgpio_probe(struct platform_device *ofdev)
 	irqmap = (s32 *)of_get_property(np, "irqmap", &size);
 	if (irqmap) {
 		if (size < gc->ngpio) {
-			dev_err(&ofdev->dev,
+			dev_err(dev,
 				"irqmap shorter than ngpio (%d < %d)\n",
 				size, gc->ngpio);
 			return -EINVAL;

@@ -389,7 +390,7 @@ static int grgpio_probe(struct platform_device *ofdev)
 						     &grgpio_irq_domain_ops,
 						     priv);
 		if (!priv->domain) {
-			dev_err(&ofdev->dev, "Could not add irq domain\n");
+			dev_err(dev, "Could not add irq domain\n");
 			return -EINVAL;
 		}

@@ -419,13 +420,13 @@ static int grgpio_probe(struct platform_device *ofdev)
 
 	err = gpiochip_add_data(gc, priv);
 	if (err) {
-		dev_err(&ofdev->dev, "Could not add gpiochip\n");
+		dev_err(dev, "Could not add gpiochip\n");
 		if (priv->domain)
 			irq_domain_remove(priv->domain);
 		return err;
 	}
 
-	dev_info(&ofdev->dev, "regs=0x%p, base=%d, ngpio=%d, irqs=%s\n",
+	dev_info(dev, "regs=0x%p, base=%d, ngpio=%d, irqs=%s\n",
 		 priv->regs, gc->base, gc->ngpio, priv->domain ? "on" : "off");
 
 	return 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #71 lore raw

PATCH 71/356 gpio: grgpio: Add NULL check in grgpio_probe

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Charles Han <[email protected]>

[ Upstream commit 050b23d081da0f29474de043e9538c1f7a351b3b ]

devm_kasprintf() can return a NULL pointer on failure,but this returned value in grgpio_probe is not checked.
Add NULL check in grgpio_probe, to handle kernel NULL pointer dereference error.

Fixes 7eb6ce2f2723 ("gpio: Convert to using %pOF instead of full_name")

Signed-off-by Charles Han <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Bartosz Golaszewski <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpio/gpio-grgpio.c | 3 +++
 1 file changed, 3 insertions(+)

drivers/gpio/gpio-grgpio.c +3 -0

index fe919d9bd46a3..637a4d45f8c77 100644
@@ -362,6 +362,9 @@ static int grgpio_probe(struct platform_device *ofdev)
 	gc->owner = THIS_MODULE;
 	gc->to_irq = grgpio_to_irq;
 	gc->label = devm_kasprintf(dev, GFP_KERNEL, "%pOF", np);
+	if (!gc->label)
+		return -ENOMEM;
+
 	gc->base = -1;
 
 	err = of_property_read_u32(np, "nbits", &prop);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #72 lore raw

PATCH 72/356 serial: amba-pl011: Use port lock wrappers

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Gleixner <[email protected]>

[ Upstream commit 68ca3e72d7463d79d29b6e4961d6028df2a88e25 ]

When a serial port is used for kernel console output, then all modifications to the UART registers which are done from other contexts, e.g. getty, termios, are interference points for the kernel console.

So far this has been ignored and the printk output is based on the principle of hope. The rework of the console infrastructure which aims to support threaded and atomic consoles, requires to mark sections which modify the UART registers as unsafe. This allows the atomic write function to make informed decisions and eventually to restore operational state. It also allows to prevent the regular UART code from modifying UART registers while printk output is in progress.

All modifications of UART registers are guarded by the UART port lock, which provides an obvious synchronization point with the console infrastructure.

To avoid adding this functionality to all UART drivers, wrap the spin_[un]lock*() invocations for uart_port::lock into helper functions which just contain the spin_[un]lock*() invocations for now. In a subsequent step these helpers will gain the console synchronization mechanisms.

Converted with coccinelle. No functional change.

Signed-off-by Thomas Gleixner <[email protected]>

Signed-off-by John Ogness <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Stable-dep-of: 2bcacc1c87ac ("serial: amba-pl011: Fix RX stall when DMA is used")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/tty/serial/amba-pl011.c | 72 ++++++++++++++++-----------------
 1 file changed, 36 insertions(+), 36 deletions(-)

drivers/tty/serial/amba-pl011.c +36 ~~-36~~

index 362bbcdece0d7..16c7703110694 100644
@@ -347,9 +347,9 @@ static int pl011_fifo_to_tty(struct uart_amba_port *uap)
 				flag = TTY_FRAME;
 		}
 
-		spin_unlock(&uap->port.lock);
+		uart_port_unlock(&uap->port);
 		sysrq = uart_handle_sysrq_char(&uap->port, ch & 255);
-		spin_lock(&uap->port.lock);
+		uart_port_lock(&uap->port);
 
 		if (!sysrq)
 			uart_insert_char(&uap->port, ch, UART011_DR_OE, ch, flag);

@@ -544,7 +544,7 @@ static void pl011_dma_tx_callback(void *data)
 	unsigned long flags;
 	u16 dmacr;
 
-	spin_lock_irqsave(&uap->port.lock, flags);
+	uart_port_lock_irqsave(&uap->port, &flags);
 	if (uap->dmatx.queued)
 		dma_unmap_single(dmatx->chan->device->dev, dmatx->dma,
 				dmatx->len, DMA_TO_DEVICE);

@@ -565,7 +565,7 @@ static void pl011_dma_tx_callback(void *data)
 	if (!(dmacr & UART011_TXDMAE) || uart_tx_stopped(&uap->port) ||
 	    uart_circ_empty(&uap->port.state->xmit)) {
 		uap->dmatx.queued = false;
-		spin_unlock_irqrestore(&uap->port.lock, flags);
+		uart_port_unlock_irqrestore(&uap->port, flags);
 		return;
 	}

@@ -576,7 +576,7 @@ static void pl011_dma_tx_callback(void *data)
 		 */
 		pl011_start_tx_pio(uap);
 
-	spin_unlock_irqrestore(&uap->port.lock, flags);
+	uart_port_unlock_irqrestore(&uap->port, flags);
 }
 
 /*

@@ -1004,7 +1004,7 @@ static void pl011_dma_rx_callback(void *data)
 	 * routine to flush out the secondary DMA buffer while
 	 * we immediately trigger the next DMA job.
 	 */
-	spin_lock_irq(&uap->port.lock);
+	uart_port_lock_irq(&uap->port);
 	/*
 	 * Rx data can be taken by the UART interrupts during
 	 * the DMA irq handler. So we check the residue here.

@@ -1020,7 +1020,7 @@ static void pl011_dma_rx_callback(void *data)
 	ret = pl011_dma_rx_trigger_dma(uap);
 
 	pl011_dma_rx_chars(uap, pending, lastbuf, false);
-	spin_unlock_irq(&uap->port.lock);
+	uart_port_unlock_irq(&uap->port);
 	/*
 	 * Do this check after we picked the DMA chars so we don't
 	 * get some IRQ immediately from RX.

@@ -1086,11 +1086,11 @@ static void pl011_dma_rx_poll(struct timer_list *t)
 	if (jiffies_to_msecs(jiffies - dmarx->last_jiffies)
 			> uap->dmarx.poll_timeout) {
 
-		spin_lock_irqsave(&uap->port.lock, flags);
+		uart_port_lock_irqsave(&uap->port, &flags);
 		pl011_dma_rx_stop(uap);
 		uap->im |= UART011_RXIM;
 		pl011_write(uap->im, uap, REG_IMSC);
-		spin_unlock_irqrestore(&uap->port.lock, flags);
+		uart_port_unlock_irqrestore(&uap->port, flags);
 
 		uap->dmarx.running = false;
 		dmaengine_terminate_all(rxchan);

@@ -1186,10 +1186,10 @@ static void pl011_dma_shutdown(struct uart_amba_port *uap)
 	while (pl011_read(uap, REG_FR) & uap->vendor->fr_busy)
 		cpu_relax();
 
-	spin_lock_irq(&uap->port.lock);
+	uart_port_lock_irq(&uap->port);
 	uap->dmacr &= ~(UART011_DMAONERR | UART011_RXDMAE | UART011_TXDMAE);
 	pl011_write(uap->dmacr, uap, REG_DMACR);
-	spin_unlock_irq(&uap->port.lock);
+	uart_port_unlock_irq(&uap->port);
 
 	if (uap->using_tx_dma) {
 		/* In theory, this should already be done by pl011_dma_flush_buffer */

@@ -1400,9 +1400,9 @@ static void pl011_throttle_rx(struct uart_port *port)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&port->lock, flags);
+	uart_port_lock_irqsave(port, &flags);
 	pl011_stop_rx(port);
-	spin_unlock_irqrestore(&port->lock, flags);
+	uart_port_unlock_irqrestore(port, flags);
 }
 
 static void pl011_enable_ms(struct uart_port *port)

@@ -1420,7 +1420,7 @@ __acquires(&uap->port.lock)
 {
 	pl011_fifo_to_tty(uap);
 
-	spin_unlock(&uap->port.lock);
+	uart_port_unlock(&uap->port);
 	tty_flip_buffer_push(&uap->port.state->port);
 	/*
 	 * If we were temporarily out of DMA mode for a while,

@@ -1445,7 +1445,7 @@ __acquires(&uap->port.lock)
 #endif
 		}
 	}
-	spin_lock(&uap->port.lock);
+	uart_port_lock(&uap->port);
 }
 
 static bool pl011_tx_char(struct uart_amba_port *uap, unsigned char c,

@@ -1551,7 +1551,7 @@ static irqreturn_t pl011_int(int irq, void *dev_id)
 	unsigned int status, pass_counter = AMBA_ISR_PASS_LIMIT;
 	int handled = 0;
 
-	spin_lock_irqsave(&uap->port.lock, flags);
+	uart_port_lock_irqsave(&uap->port, &flags);
 	status = pl011_read(uap, REG_RIS) & uap->im;
 	if (status) {
 		do {

@@ -1581,7 +1581,7 @@ static irqreturn_t pl011_int(int irq, void *dev_id)
 		handled = 1;
 	}
 
-	spin_unlock_irqrestore(&uap->port.lock, flags);
+	uart_port_unlock_irqrestore(&uap->port, flags);
 
 	return IRQ_RETVAL(handled);
 }

@@ -1653,14 +1653,14 @@ static void pl011_break_ctl(struct uart_port *port, int break_state)
 	unsigned long flags;
 	unsigned int lcr_h;
 
-	spin_lock_irqsave(&uap->port.lock, flags);
+	uart_port_lock_irqsave(&uap->port, &flags);
 	lcr_h = pl011_read(uap, REG_LCRH_TX);
 	if (break_state == -1)
 		lcr_h |= UART01x_LCRH_BRK;
 	else
 		lcr_h &= ~UART01x_LCRH_BRK;
 	pl011_write(lcr_h, uap, REG_LCRH_TX);
-	spin_unlock_irqrestore(&uap->port.lock, flags);
+	uart_port_unlock_irqrestore(&uap->port, flags);
 }
 
 #ifdef CONFIG_CONSOLE_POLL

@@ -1799,7 +1799,7 @@ static void pl011_enable_interrupts(struct uart_amba_port *uap)
 	unsigned long flags;
 	unsigned int i;
 
-	spin_lock_irqsave(&uap->port.lock, flags);
+	uart_port_lock_irqsave(&uap->port, &flags);
 
 	/* Clear out any spuriously appearing RX interrupts */
 	pl011_write(UART011_RTIS | UART011_RXIS, uap, REG_ICR);

@@ -1821,7 +1821,7 @@ static void pl011_enable_interrupts(struct uart_amba_port *uap)
 	if (!pl011_dma_rx_running(uap))
 		uap->im |= UART011_RXIM;
 	pl011_write(uap->im, uap, REG_IMSC);
-	spin_unlock_irqrestore(&uap->port.lock, flags);
+	uart_port_unlock_irqrestore(&uap->port, flags);
 }
 
 static void pl011_unthrottle_rx(struct uart_port *port)

@@ -1829,7 +1829,7 @@ static void pl011_unthrottle_rx(struct uart_port *port)
 	struct uart_amba_port *uap = container_of(port, struct uart_amba_port, port);
 	unsigned long flags;
 
-	spin_lock_irqsave(&uap->port.lock, flags);
+	uart_port_lock_irqsave(&uap->port, &flags);
 
 	uap->im = UART011_RTIM;
 	if (!pl011_dma_rx_running(uap))

@@ -1837,7 +1837,7 @@ static void pl011_unthrottle_rx(struct uart_port *port)
 
 	pl011_write(uap->im, uap, REG_IMSC);
 
-	spin_unlock_irqrestore(&uap->port.lock, flags);
+	uart_port_unlock_irqrestore(&uap->port, flags);
 }
 
 static int pl011_startup(struct uart_port *port)

@@ -1857,7 +1857,7 @@ static int pl011_startup(struct uart_port *port)
 
 	pl011_write(uap->vendor->ifls, uap, REG_IFLS);
 
-	spin_lock_irq(&uap->port.lock);
+	uart_port_lock_irq(&uap->port);
 
 	cr = pl011_read(uap, REG_CR);
 	cr &= UART011_CR_RTS | UART011_CR_DTR;

@@ -1868,7 +1868,7 @@ static int pl011_startup(struct uart_port *port)
 
 	pl011_write(cr, uap, REG_CR);
 
-	spin_unlock_irq(&uap->port.lock);
+	uart_port_unlock_irq(&uap->port);
 
 	/*
 	 * initialise the old status of the modem signals

@@ -1929,12 +1929,12 @@ static void pl011_disable_uart(struct uart_amba_port *uap)
 	unsigned int cr;
 
 	uap->port.status &= ~(UPSTAT_AUTOCTS | UPSTAT_AUTORTS);
-	spin_lock_irq(&uap->port.lock);
+	uart_port_lock_irq(&uap->port);
 	cr = pl011_read(uap, REG_CR);
 	cr &= UART011_CR_RTS | UART011_CR_DTR;
 	cr |= UART01x_CR_UARTEN | UART011_CR_TXE;
 	pl011_write(cr, uap, REG_CR);
-	spin_unlock_irq(&uap->port.lock);
+	uart_port_unlock_irq(&uap->port);
 
 	/*
 	 * disable break condition and fifos

@@ -1946,14 +1946,14 @@ static void pl011_disable_uart(struct uart_amba_port *uap)
 
 static void pl011_disable_interrupts(struct uart_amba_port *uap)
 {
-	spin_lock_irq(&uap->port.lock);
+	uart_port_lock_irq(&uap->port);
 
 	/* mask all interrupts and clear all pending ones */
 	uap->im = 0;
 	pl011_write(uap->im, uap, REG_IMSC);
 	pl011_write(0xffff, uap, REG_ICR);
 
-	spin_unlock_irq(&uap->port.lock);
+	uart_port_unlock_irq(&uap->port);
 }
 
 static void pl011_shutdown(struct uart_port *port)

@@ -2098,7 +2098,7 @@ pl011_set_termios(struct uart_port *port, struct ktermios *termios,
 
 	bits = tty_get_frame_size(termios->c_cflag);
 
-	spin_lock_irqsave(&port->lock, flags);
+	uart_port_lock_irqsave(port, &flags);
 
 	/*
 	 * Update the per-port timeout.

@@ -2172,7 +2172,7 @@ pl011_set_termios(struct uart_port *port, struct ktermios *termios,
 	old_cr |= UART011_CR_RXE;
 	pl011_write(old_cr, uap, REG_CR);
 
-	spin_unlock_irqrestore(&port->lock, flags);
+	uart_port_unlock_irqrestore(port, flags);
 }
 
 static void

@@ -2190,10 +2190,10 @@ sbsa_uart_set_termios(struct uart_port *port, struct ktermios *termios,
 	termios->c_cflag &= ~(CMSPAR | CRTSCTS);
 	termios->c_cflag |= CS8 | CLOCAL;
 
-	spin_lock_irqsave(&port->lock, flags);
+	uart_port_lock_irqsave(port, &flags);
 	uart_update_timeout(port, CS8, uap->fixed_baud);
 	pl011_setup_status_masks(port, termios);
-	spin_unlock_irqrestore(&port->lock, flags);
+	uart_port_unlock_irqrestore(port, flags);
 }
 
 static const char *pl011_type(struct uart_port *port)

@@ -2332,9 +2332,9 @@ pl011_console_write(struct console *co, const char *s, unsigned int count)
 	if (uap->port.sysrq)
 		locked = 0;
 	else if (oops_in_progress)
-		locked = spin_trylock(&uap->port.lock);
+		locked = uart_port_trylock(&uap->port);
 	else
-		spin_lock(&uap->port.lock);
+		uart_port_lock(&uap->port);
 
 	/*
 	 *	First save the CR then disable the interrupts

@@ -2360,7 +2360,7 @@ pl011_console_write(struct console *co, const char *s, unsigned int count)
 		pl011_write(old_cr, uap, REG_CR);
 
 	if (locked)
-		spin_unlock(&uap->port.lock);
+		uart_port_unlock(&uap->port);
 	local_irq_restore(flags);
 
 	clk_disable(uap->clk);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #73 lore raw

PATCH 73/356 serial: amba-pl011: Fix RX stall when DMA is used

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kartik Rajput <[email protected]>

[ Upstream commit 2bcacc1c87acf9a8ebc17de18cb2b3cfeca547cf ]

Function pl011_throttle_rx() calls pl011_stop_rx() to disable RX, which also disables the RX DMA by clearing the RXDMAE bit of the DMACR register. However, to properly unthrottle RX when DMA is used, the function pl011_unthrottle_rx() is expected to set the RXDMAE bit of the DMACR register, which it currently lacks. This causes RX to stall after the throttle API is called.

Set RXDMAE bit in the DMACR register while unthrottling RX if RX DMA is used.

Fixes 211565b10099 ("serial: pl011: UPSTAT_AUTORTS requires .throttle/unthrottle")

Signed-off-by Kartik Rajput <[email protected]>

Reviewed-by Linus Walleij <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/tty/serial/amba-pl011.c | 5 +++++
 1 file changed, 5 insertions(+)

drivers/tty/serial/amba-pl011.c +5 -0

index 16c7703110694..08f80188f73dd 100644
@@ -1837,6 +1837,11 @@ static void pl011_unthrottle_rx(struct uart_port *port)
 
 	pl011_write(uap->im, uap, REG_IMSC);
 
+	if (uap->using_rx_dma) {
+		uap->dmacr |= UART011_RXDMAE;
+		pl011_write(uap->dmacr, uap, REG_DMACR);
+	}
+
 	uart_port_unlock_irqrestore(&uap->port, flags);
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #74 lore raw

PATCH 74/356 soc/fsl: cpm: qmc: Convert to platform remove callback returning void

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Uwe Kleine-König <[email protected]>

[ Upstream commit 14914a115e807aa2f8025e451133627a64120ac3 ]

The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new() which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove().

Trivially convert this driver from always returning zero in the remove callback to the void returning variant.

Acked-by Herve Codina <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Uwe Kleine-König <[email protected]>

Stable-dep-of: cb3daa51db81 ("soc: fsl: cpm1: qmc: Set the ret error code on platform_get_irq() failure")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/soc/fsl/qe/qmc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

drivers/soc/fsl/qe/qmc.c +2 -4

index 8dc73cc1a83b1..2312152a44b3e 100644
@@ -1414,7 +1414,7 @@ static int qmc_probe(struct platform_device *pdev)
 	return ret;
 }
 
-static int qmc_remove(struct platform_device *pdev)
+static void qmc_remove(struct platform_device *pdev)
 {
 	struct qmc *qmc = platform_get_drvdata(pdev);

@@ -1426,8 +1426,6 @@ static int qmc_remove(struct platform_device *pdev)
 
 	/* Disconnect the serial from TSA */
 	tsa_serial_disconnect(qmc->tsa_serial);
-
-	return 0;
 }
 
 static const struct of_device_id qmc_id_table[] = {

@@ -1442,7 +1440,7 @@ static struct platform_driver qmc_driver = {
 		.of_match_table = of_match_ptr(qmc_id_table),
 	},
 	.probe = qmc_probe,
-	.remove = qmc_remove,
+	.remove_new = qmc_remove,
 };
 module_platform_driver(qmc_driver);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #75 lore raw

PATCH 75/356 soc: fsl: cpm1: qmc: Fix blank line and spaces

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Herve Codina <[email protected]>

[ Upstream commit f06ab938bcddcb3c3a0b458b03a827c701919c9e ]

checkpatch.pl raises the following issues
CHECK: Please don't use multiple blank lines
CHECK: Alignment should match open parenthesis

Fix them.

Link https://lore.kernel.org/r/[email protected] open here

Stable-dep-of: cb3daa51db81 ("soc: fsl: cpm1: qmc: Set the ret error code on platform_get_irq() failure")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/soc/fsl/qe/qmc.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

drivers/soc/fsl/qe/qmc.c +4 -6

index 2312152a44b3e..f22d1d85d1021 100644
@@ -253,7 +253,6 @@ static inline void qmc_setbits32(void __iomem *addr, u32 set)
 	qmc_write32(addr, qmc_read32(addr) | set);
 }
 
-
 int qmc_chan_get_info(struct qmc_chan *chan, struct qmc_chan_info *info)
 {
 	struct tsa_serial_info tsa_info;

@@ -1093,7 +1092,7 @@ static int qmc_setup_chan(struct qmc *qmc, struct qmc_chan *chan)
 		qmc_write32(chan->s_param + QMC_SPE_ZDSTATE, 0x00000080);
 		qmc_write16(chan->s_param + QMC_SPE_MFLR, 60);
 		qmc_write16(chan->s_param + QMC_SPE_CHAMR,
-			QMC_SPE_CHAMR_MODE_HDLC | QMC_SPE_CHAMR_HDLC_IDLM);
+			    QMC_SPE_CHAMR_MODE_HDLC | QMC_SPE_CHAMR_HDLC_IDLM);
 	}
 
 	/* Do not enable interrupts now. They will be enabled later */

@@ -1286,7 +1285,6 @@ static int qmc_probe(struct platform_device *pdev)
 	if (IS_ERR(qmc->scc_regs))
 		return PTR_ERR(qmc->scc_regs);
 
-
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "scc_pram");
 	if (!res)
 		return -EINVAL;

@@ -1332,7 +1330,7 @@ static int qmc_probe(struct platform_device *pdev)
 	 */
 	qmc->bd_size = (nb_chans * (QMC_NB_TXBDS + QMC_NB_RXBDS)) * sizeof(cbd_t);
 	qmc->bd_table = dmam_alloc_coherent(qmc->dev, qmc->bd_size,
-		&qmc->bd_dma_addr, GFP_KERNEL);
+					    &qmc->bd_dma_addr, GFP_KERNEL);
 	if (!qmc->bd_table) {
 		dev_err(qmc->dev, "Failed to allocate bd table\n");
 		ret = -ENOMEM;

@@ -1345,7 +1343,7 @@ static int qmc_probe(struct platform_device *pdev)
 	/* Allocate the interrupt table */
 	qmc->int_size = QMC_NB_INTS * sizeof(u16);
 	qmc->int_table = dmam_alloc_coherent(qmc->dev, qmc->int_size,
-		&qmc->int_dma_addr, GFP_KERNEL);
+					     &qmc->int_dma_addr, GFP_KERNEL);
 	if (!qmc->int_table) {
 		dev_err(qmc->dev, "Failed to allocate interrupt table\n");
 		ret = -ENOMEM;

@@ -1393,7 +1391,7 @@ static int qmc_probe(struct platform_device *pdev)
 
 	/* Enable interrupts */
 	qmc_write16(qmc->scc_regs + SCC_SCCM,
-		SCC_SCCE_IQOV | SCC_SCCE_GINT | SCC_SCCE_GUN | SCC_SCCE_GOV);
+		    SCC_SCCE_IQOV | SCC_SCCE_GINT | SCC_SCCE_GUN | SCC_SCCE_GOV);
 
 	ret = qmc_finalize_chans(qmc);
 	if (ret < 0)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #76 lore raw

PATCH 76/356 soc: fsl: cpm1: qmc: Re-order probe() operations

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Herve Codina <[email protected]>

[ Upstream commit a13bf605342ea9df492b8159cadaa41862b53e15 ]

Current code handles CPM1 version of QMC. In the QUICC Engine (QE) version, some operations done at probe() need to be done in a different order.

In order to prepare the support for the QE version, changed the sequence of operation done at probe():
- Retrieve the tsa_serial earlier, before initializing resources.
- Group SCC initialisation and do this initialization when it is really needed in the probe() sequence.

Having the QE compatible sequence in the CPM1 version does not lead to any issue and works correctly without any regressions.

Link https://lore.kernel.org/r/[email protected] open here

Stable-dep-of: cb3daa51db81 ("soc: fsl: cpm1: qmc: Set the ret error code on platform_get_irq() failure")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/soc/fsl/qe/qmc.c | 54 +++++++++++++++++++---------------------
 1 file changed, 26 insertions(+), 28 deletions(-)

drivers/soc/fsl/qe/qmc.c +26 ~~-28~~

index f22d1d85d1021..a5c9cbb99600e 100644
@@ -1281,6 +1281,12 @@ static int qmc_probe(struct platform_device *pdev)
 	qmc->dev = &pdev->dev;
 	INIT_LIST_HEAD(&qmc->chan_head);
 
+	qmc->tsa_serial = devm_tsa_serial_get_byphandle(qmc->dev, np, "fsl,tsa-serial");
+	if (IS_ERR(qmc->tsa_serial)) {
+		return dev_err_probe(qmc->dev, PTR_ERR(qmc->tsa_serial),
+				     "Failed to get TSA serial\n");
+	}
+
 	qmc->scc_regs = devm_platform_ioremap_resource_byname(pdev, "scc_regs");
 	if (IS_ERR(qmc->scc_regs))
 		return PTR_ERR(qmc->scc_regs);

@@ -1297,33 +1303,13 @@ static int qmc_probe(struct platform_device *pdev)
 	if (IS_ERR(qmc->dpram))
 		return PTR_ERR(qmc->dpram);
 
-	qmc->tsa_serial = devm_tsa_serial_get_byphandle(qmc->dev, np, "fsl,tsa-serial");
-	if (IS_ERR(qmc->tsa_serial)) {
-		return dev_err_probe(qmc->dev, PTR_ERR(qmc->tsa_serial),
-				     "Failed to get TSA serial\n");
-	}
-
-	/* Connect the serial (SCC) to TSA */
-	ret = tsa_serial_connect(qmc->tsa_serial);
-	if (ret) {
-		dev_err(qmc->dev, "Failed to connect TSA serial\n");
-		return ret;
-	}
-
 	/* Parse channels informationss */
 	ret = qmc_of_parse_chans(qmc, np);
 	if (ret)
-		goto err_tsa_serial_disconnect;
+		return ret;
 
 	nb_chans = qmc_nb_chans(qmc);
 
-	/* Init GMSR_H and GMSR_L registers */
-	qmc_write32(qmc->scc_regs + SCC_GSMRH,
-		    SCC_GSMRH_CDS | SCC_GSMRH_CTSS | SCC_GSMRH_CDP | SCC_GSMRH_CTSP);
-
-	/* enable QMC mode */
-	qmc_write32(qmc->scc_regs + SCC_GSMRL, SCC_GSMRL_MODE_QMC);
-
 	/*
 	 * Allocate the buffer descriptor table
 	 * 8 rx and 8 tx descriptors per channel

@@ -1333,8 +1319,7 @@ static int qmc_probe(struct platform_device *pdev)
 					    &qmc->bd_dma_addr, GFP_KERNEL);
 	if (!qmc->bd_table) {
 		dev_err(qmc->dev, "Failed to allocate bd table\n");
-		ret = -ENOMEM;
-		goto err_tsa_serial_disconnect;
+		return -ENOMEM;
 	}
 	memset(qmc->bd_table, 0, qmc->bd_size);

@@ -1346,8 +1331,7 @@ static int qmc_probe(struct platform_device *pdev)
 					     &qmc->int_dma_addr, GFP_KERNEL);
 	if (!qmc->int_table) {
 		dev_err(qmc->dev, "Failed to allocate interrupt table\n");
-		ret = -ENOMEM;
-		goto err_tsa_serial_disconnect;
+		return -ENOMEM;
 	}
 	memset(qmc->int_table, 0, qmc->int_size);

@@ -1366,18 +1350,32 @@ static int qmc_probe(struct platform_device *pdev)
 
 	ret = qmc_setup_tsa(qmc);
 	if (ret)
-		goto err_tsa_serial_disconnect;
+		return ret;
 
 	qmc_write16(qmc->scc_pram + QMC_GBL_QMCSTATE, 0x8000);
 
 	ret = qmc_setup_chans(qmc);
 	if (ret)
-		goto err_tsa_serial_disconnect;
+		return ret;
 
 	/* Init interrupts table */
 	ret = qmc_setup_ints(qmc);
 	if (ret)
-		goto err_tsa_serial_disconnect;
+		return ret;
+
+	/* Connect the serial (SCC) to TSA */
+	ret = tsa_serial_connect(qmc->tsa_serial);
+	if (ret) {
+		dev_err(qmc->dev, "Failed to connect TSA serial\n");
+		return ret;
+	}
+
+	/* Init GMSR_H and GMSR_L registers */
+	qmc_write32(qmc->scc_regs + SCC_GSMRH,
+		    SCC_GSMRH_CDS | SCC_GSMRH_CTSS | SCC_GSMRH_CDP | SCC_GSMRH_CTSP);
+
+	/* enable QMC mode */
+	qmc_write32(qmc->scc_regs + SCC_GSMRL, SCC_GSMRL_MODE_QMC);
 
 	/* Disable and clear interrupts,  set the irq handler */
 	qmc_write16(qmc->scc_regs + SCC_SCCM, 0x0000);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #77 lore raw

PATCH 77/356 soc: fsl: cpm1: qmc: Introduce qmc_init_resource() and its CPM1 version

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Herve Codina <[email protected]>

[ Upstream commit 727b3ab490a5f5e74fb3f246c9fdfb339d309950 ]

Current code handles the CPM1 version of QMC. Resources initialisations (i.e. retrieving base addresses and offsets of different parts) will be slightly different in the QUICC Engine (QE) version. Indeed, in QE version, some resources need to be allocated and are no more "staticaly" defined.

In order to prepare the support for QE version, introduce qmc_init_resource() to initialize those resources and isolate the CPM1 specific operations in a specific function.

Link https://lore.kernel.org/r/[email protected] open here

Stable-dep-of: cb3daa51db81 ("soc: fsl: cpm1: qmc: Set the ret error code on platform_get_irq() failure")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/soc/fsl/qe/qmc.c | 47 ++++++++++++++++++++++++++--------------
 1 file changed, 31 insertions(+), 16 deletions(-)

drivers/soc/fsl/qe/qmc.c +31 ~~-16~~

index a5c9cbb99600e..f2bda8658e034 100644
@@ -1265,11 +1265,38 @@ static irqreturn_t qmc_irq_handler(int irq, void *priv)
 	return IRQ_HANDLED;
 }
 
+static int qmc_cpm1_init_resources(struct qmc *qmc, struct platform_device *pdev)
+{
+	struct resource *res;
+
+	qmc->scc_regs = devm_platform_ioremap_resource_byname(pdev, "scc_regs");
+	if (IS_ERR(qmc->scc_regs))
+		return PTR_ERR(qmc->scc_regs);
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "scc_pram");
+	if (!res)
+		return -EINVAL;
+	qmc->scc_pram_offset = res->start - get_immrbase();
+	qmc->scc_pram = devm_ioremap_resource(qmc->dev, res);
+	if (IS_ERR(qmc->scc_pram))
+		return PTR_ERR(qmc->scc_pram);
+
+	qmc->dpram  = devm_platform_ioremap_resource_byname(pdev, "dpram");
+	if (IS_ERR(qmc->dpram))
+		return PTR_ERR(qmc->dpram);
+
+	return 0;
+}
+
+static int qmc_init_resources(struct qmc *qmc, struct platform_device *pdev)
+{
+	return qmc_cpm1_init_resources(qmc, pdev);
+}
+
 static int qmc_probe(struct platform_device *pdev)
 {
 	struct device_node *np = pdev->dev.of_node;
 	unsigned int nb_chans;
-	struct resource *res;
 	struct qmc *qmc;
 	int irq;
 	int ret;

@@ -1287,21 +1314,9 @@ static int qmc_probe(struct platform_device *pdev)
 				     "Failed to get TSA serial\n");
 	}
 
-	qmc->scc_regs = devm_platform_ioremap_resource_byname(pdev, "scc_regs");
-	if (IS_ERR(qmc->scc_regs))
-		return PTR_ERR(qmc->scc_regs);
-
-	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "scc_pram");
-	if (!res)
-		return -EINVAL;
-	qmc->scc_pram_offset = res->start - get_immrbase();
-	qmc->scc_pram = devm_ioremap_resource(qmc->dev, res);
-	if (IS_ERR(qmc->scc_pram))
-		return PTR_ERR(qmc->scc_pram);
-
-	qmc->dpram  = devm_platform_ioremap_resource_byname(pdev, "dpram");
-	if (IS_ERR(qmc->dpram))
-		return PTR_ERR(qmc->dpram);
+	ret = qmc_init_resources(qmc, pdev);
+	if (ret)
+		return ret;
 
 	/* Parse channels informationss */
 	ret = qmc_of_parse_chans(qmc, np);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #78 lore raw

PATCH 78/356 soc: fsl: cpm1: qmc: Introduce qmc_{init,exit}_xcc() and their CPM1 version

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Herve Codina <[email protected]>

[ Upstream commit de5fdb7d14b34f7fea930f2d72cf0241ec679e72 ]

Current code handles the CPM1 version of QMC and initialize the QMC used SCC. The QUICC Engine (QE) version uses an UCC (Unified Communication Controllers) instead of the SCC (Serial Communication Controllers) used in the CPM1 version. These controllers serve the same purpose and are used in the same way but their inializations are slightly different.

In order to prepare the support for QE version of QMC, introduce qmc_init_xcc() to initialize theses controllers (UCC in QE and SCC in CPM1) and isolate the CPM1 specific SCC initialization in a specific function.

Also introduce qmc_exit_xcc() for consistency to revert operations done in qmc_init_xcc().

Link https://lore.kernel.org/r/[email protected] open here

Stable-dep-of: cb3daa51db81 ("soc: fsl: cpm1: qmc: Set the ret error code on platform_get_irq() failure")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/soc/fsl/qe/qmc.c | 66 +++++++++++++++++++++++++++-------------
 1 file changed, 45 insertions(+), 21 deletions(-)

drivers/soc/fsl/qe/qmc.c +45 ~~-21~~

index f2bda8658e034..9fa75effcfc06 100644
@@ -1293,6 +1293,41 @@ static int qmc_init_resources(struct qmc *qmc, struct platform_device *pdev)
 	return qmc_cpm1_init_resources(qmc, pdev);
 }
 
+static int qmc_cpm1_init_scc(struct qmc *qmc)
+{
+	u32 val;
+	int ret;
+
+	/* Connect the serial (SCC) to TSA */
+	ret = tsa_serial_connect(qmc->tsa_serial);
+	if (ret)
+		return dev_err_probe(qmc->dev, ret, "Failed to connect TSA serial\n");
+
+	/* Init GMSR_H and GMSR_L registers */
+	val = SCC_GSMRH_CDS | SCC_GSMRH_CTSS | SCC_GSMRH_CDP | SCC_GSMRH_CTSP;
+	qmc_write32(qmc->scc_regs + SCC_GSMRH, val);
+
+	/* enable QMC mode */
+	qmc_write32(qmc->scc_regs + SCC_GSMRL, SCC_GSMRL_MODE_QMC);
+
+	/* Disable and clear interrupts */
+	qmc_write16(qmc->scc_regs + SCC_SCCM, 0x0000);
+	qmc_write16(qmc->scc_regs + SCC_SCCE, 0x000F);
+
+	return 0;
+}
+
+static int qmc_init_xcc(struct qmc *qmc)
+{
+	return qmc_cpm1_init_scc(qmc);
+}
+
+static void qmc_exit_xcc(struct qmc *qmc)
+{
+	/* Disconnect the serial from TSA */
+	tsa_serial_disconnect(qmc->tsa_serial);
+}
+
 static int qmc_probe(struct platform_device *pdev)
 {
 	struct device_node *np = pdev->dev.of_node;

@@ -1378,29 +1413,18 @@ static int qmc_probe(struct platform_device *pdev)
 	if (ret)
 		return ret;
 
-	/* Connect the serial (SCC) to TSA */
-	ret = tsa_serial_connect(qmc->tsa_serial);
-	if (ret) {
-		dev_err(qmc->dev, "Failed to connect TSA serial\n");
+	/* Init SCC */
+	ret = qmc_init_xcc(qmc);
+	if (ret)
 		return ret;
-	}
 
-	/* Init GMSR_H and GMSR_L registers */
-	qmc_write32(qmc->scc_regs + SCC_GSMRH,
-		    SCC_GSMRH_CDS | SCC_GSMRH_CTSS | SCC_GSMRH_CDP | SCC_GSMRH_CTSP);
-
-	/* enable QMC mode */
-	qmc_write32(qmc->scc_regs + SCC_GSMRL, SCC_GSMRL_MODE_QMC);
-
-	/* Disable and clear interrupts,  set the irq handler */
-	qmc_write16(qmc->scc_regs + SCC_SCCM, 0x0000);
-	qmc_write16(qmc->scc_regs + SCC_SCCE, 0x000F);
+	/* Set the irq handler */
 	irq = platform_get_irq(pdev, 0);
 	if (irq < 0)
-		goto err_tsa_serial_disconnect;
+		goto err_exit_xcc;
 	ret = devm_request_irq(qmc->dev, irq, qmc_irq_handler, 0, "qmc", qmc);
 	if (ret < 0)
-		goto err_tsa_serial_disconnect;
+		goto err_exit_xcc;
 
 	/* Enable interrupts */
 	qmc_write16(qmc->scc_regs + SCC_SCCM,

@@ -1420,8 +1444,8 @@ static int qmc_probe(struct platform_device *pdev)
 err_disable_intr:
 	qmc_write16(qmc->scc_regs + SCC_SCCM, 0);
 
-err_tsa_serial_disconnect:
-	tsa_serial_disconnect(qmc->tsa_serial);
+err_exit_xcc:
+	qmc_exit_xcc(qmc);
 	return ret;
 }

@@ -1435,8 +1459,8 @@ static void qmc_remove(struct platform_device *pdev)
 	/* Disable interrupts */
 	qmc_write16(qmc->scc_regs + SCC_SCCM, 0);
 
-	/* Disconnect the serial from TSA */
-	tsa_serial_disconnect(qmc->tsa_serial);
+	/* Exit SCC */
+	qmc_exit_xcc(qmc);
 }
 
 static const struct of_device_id qmc_id_table[] = {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #79 lore raw

PATCH 79/356 soc: fsl: cpm1: qmc: Set the ret error code on platform_get_irq() failure

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Herve Codina <[email protected]>

[ Upstream commit cb3daa51db819a172e9524e96e2ed96b4237e51a ]

A kernel test robot detected a missing error code:
qmc.c:1942 qmc_probe() warn: missing error code 'ret'

Indeed, the error returned by platform_get_irq() is checked and the operation is aborted in case of failure but the ret error code is not set in that case.

Set the ret error code.

Reported-by kernel test robot <[email protected]>

Reported-by Dan Carpenter <[email protected]>

Closes https://lore.kernel.org/r/[email protected]/ open here

Fixes 3178d58e0b97 ("soc: fsl: cpm1: Add support for QMC")

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/soc/fsl/qe/qmc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

drivers/soc/fsl/qe/qmc.c +3 -1

index 9fa75effcfc06..f1720c7cbe063 100644
@@ -1420,8 +1420,10 @@ static int qmc_probe(struct platform_device *pdev)
 
 	/* Set the irq handler */
 	irq = platform_get_irq(pdev, 0);
-	if (irq < 0)
+	if (irq < 0) {
+		ret = irq;
 		goto err_exit_xcc;
+	}
 	ret = devm_request_irq(qmc->dev, irq, qmc_irq_handler, 0, "qmc", qmc);
 	if (ret < 0)
 		goto err_exit_xcc;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #80 lore raw

PATCH 80/356 usb: dwc3: gadget: Rewrite endpoint allocation flow

Link https://lore.kernel.org/r/c143583a5afb087deb8c3aa5eb227ee23515f272.1706754219.git.Thinh.Nguyen@synopsys.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thinh Nguyen <[email protected]>

[ Upstream commit b311048c174da893f47fc09439bc1f6fa2a29589 ]

The driver dwc3 deviates from the programming guide in regard to endpoint configuration. It does this command sequence:

DEPSTARTCFG -> DEPXFERCFG -> DEPCFG

Instead of the suggested flow:

DEPSTARTCFG -> DEPCFG -> DEPXFERCFG

The reasons for this deviation were as follow, quoted:

1) The databook says to do %DWC3_DEPCMD_DEPSTARTCFG for every %USB_REQ_SET_CONFIGURATION and %USB_REQ_SET_INTERFACE (8.1.5). This is incorrect in the scenario of multiple interfaces.

2) The databook does not mention doing more %DWC3_DEPCMD_DEPXFERCFG for new endpoint on alt setting (8.1.6).

Regarding 1), DEPSTARTCFG resets the endpoints' resource and can be a problem if used with SET_INTERFACE request of a multiple interface configuration. But we can still satisfy the programming guide requirement by assigning the endpoint resource as part of usb_ep_enable(). We will only reset endpoint resources on controller initialization and SET_CONFIGURATION request.

Regarding 2), the later versions of the programming guide were updated to clarify this flow (see "Alternate Initialization on SetInterface Request" of the programming guide). As long as the platform has enough physical endpoints, we can assign resource to a new endpoint.

The order of the command sequence will not be a problem to most platforms for the current implementation of the dwc3 driver. However, this order is required in different scenarios (such as initialization during controller's hibernation restore). Let's keep the flow consistent and follow the programming guide.

Signed-off-by Thinh Nguyen <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

Stable-dep-of: 5d2fb074dea2 ("usb: dwc3: ep0: Don't clear ep0 DWC3_EP_TRANSFER_STARTED")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/usb/dwc3/core.h   |  1 +
 drivers/usb/dwc3/ep0.c    |  1 +
 drivers/usb/dwc3/gadget.c | 89 +++++++++++++++++----------------------
 drivers/usb/dwc3/gadget.h |  1 +
 4 files changed, 41 insertions(+), 51 deletions(-)

drivers/usb/dwc3/core.h +1 -0

index 3325796f3cb45..b118f4aab1898 100644
@@ -751,6 +751,7 @@ struct dwc3_ep {
 #define DWC3_EP_PENDING_CLEAR_STALL	BIT(11)
 #define DWC3_EP_TXFIFO_RESIZED		BIT(12)
 #define DWC3_EP_DELAY_STOP             BIT(13)
+#define DWC3_EP_RESOURCE_ALLOCATED	BIT(14)
 
 	/* This last one is specific to EP0 */
 #define DWC3_EP0_DIR_IN			BIT(31)

drivers/usb/dwc3/ep0.c +1 -0

index 6ae8a36f21cf6..72bb722da2f25 100644
@@ -646,6 +646,7 @@ static int dwc3_ep0_set_config(struct dwc3 *dwc, struct usb_ctrlrequest *ctrl)
 		return -EINVAL;
 
 	case USB_STATE_ADDRESS:
+		dwc3_gadget_start_config(dwc, 2);
 		dwc3_gadget_clear_tx_fifos(dwc);
 
 		ret = dwc3_ep0_delegate_req(dwc, ctrl);

drivers/usb/dwc3/gadget.c +38 ~~-51~~

144 lines not on this page · raw message

drivers/usb/dwc3/gadget.h +1 -0

index 55a56cf67d736..d73e735e40810 100644
@@ -119,6 +119,7 @@ int dwc3_gadget_ep0_queue(struct usb_ep *ep, struct usb_request *request,
 int __dwc3_gadget_ep_set_halt(struct dwc3_ep *dep, int value, int protocol);
 void dwc3_ep0_send_delayed_status(struct dwc3 *dwc);
 void dwc3_stop_active_transfer(struct dwc3_ep *dep, bool force, bool interrupt);
+int dwc3_gadget_start_config(struct dwc3 *dwc, unsigned int resource_index);
 
 /**
  * dwc3_gadget_ep_get_transfer_index - Gets transfer index from HW

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #81 lore raw

PATCH 81/356 usb: dwc3: ep0: Dont reset resource alloc flag (including ep0)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Michael Grzeschik <[email protected]>

[ Upstream commit 72fca8371f205d654f95b09cd023a71fd5307041 ]

The DWC3_EP_RESOURCE_ALLOCATED flag ensures that the resource of an endpoint is only assigned once. Unless the endpoint is reset, don't clear this flag. Otherwise we may set endpoint resource again, which prevents the driver from initiate transfer after handling a STALL or endpoint halt to the control endpoint.

Commit f2e0eee47038 ("usb: dwc3: ep0: Don't reset resource alloc flag") was fixing the initial issue, but did this only for physical ep1. Since the function dwc3_ep0_stall_and_restart is resetting the flags for both physical endpoints, this also has to be done for ep0.

Fixes b311048c174d ("usb: dwc3: gadget: Rewrite endpoint allocation flow")

Acked-by Thinh Nguyen <[email protected]>

Signed-off-by Michael Grzeschik <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Stable-dep-of: 5d2fb074dea2 ("usb: dwc3: ep0: Don't clear ep0 DWC3_EP_TRANSFER_STARTED")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/usb/dwc3/ep0.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/usb/dwc3/ep0.c +2 -1

index 72bb722da2f25..371662a552538 100644
@@ -231,7 +231,8 @@ void dwc3_ep0_stall_and_restart(struct dwc3 *dwc)
 	/* stall is always issued on EP0 */
 	dep = dwc->eps[0];
 	__dwc3_gadget_ep_set_halt(dep, 1, false);
-	dep->flags = DWC3_EP_ENABLED;
+	dep->flags &= DWC3_EP_RESOURCE_ALLOCATED;
+	dep->flags |= DWC3_EP_ENABLED;
 	dwc->delayed_status = false;
 
 	if (!list_empty(&dep->pending_list)) {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #82 lore raw

PATCH 82/356 usb: dwc3: ep0: Dont clear ep0 DWC3_EP_TRANSFER_STARTED

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thinh Nguyen <[email protected]>

[ Upstream commit 5d2fb074dea289c41f5aaf2c3f68286bee370634 ]

The driver cannot issue the End Transfer command to the SETUP transfer. Don't clear DWC3_EP_TRANSFER_STARTED flag to make sure that the driver won't send Start Transfer command again, which can cause no-resource error. For example this can occur if the host issues a reset to the device.

Link https://lore.kernel.org/r/d3d618185fd614bb7426352a9fc1199641d3b5f5.1731545781.git.Thinh.Nguyen@synopsys.com open here

Fixes 76cb323f80ac ("usb: dwc3: ep0: clear all EP0 flags")

Signed-off-by Thinh Nguyen <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/usb/dwc3/ep0.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/usb/dwc3/ep0.c +1 -1

index 371662a552538..69d98f1938edd 100644
@@ -231,7 +231,7 @@ void dwc3_ep0_stall_and_restart(struct dwc3 *dwc)
 	/* stall is always issued on EP0 */
 	dep = dwc->eps[0];
 	__dwc3_gadget_ep_set_halt(dep, 1, false);
-	dep->flags &= DWC3_EP_RESOURCE_ALLOCATED;
+	dep->flags &= DWC3_EP_RESOURCE_ALLOCATED | DWC3_EP_TRANSFER_STARTED;
 	dep->flags |= DWC3_EP_ENABLED;
 	dwc->delayed_status = false;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #83 lore raw

PATCH 83/356 mmc: mtk-sd: use devm_mmc_alloc_host

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rosen Penev <[email protected]>

[ Upstream commit 7a2fa8eed936b33b22e49b1d2349cd7d02f22710 ]

Allows removing several gotos.

Also fixed some wrong ones.

Added dev_err_probe where EPROBE_DEFER is possible.

Signed-off-by Rosen Penev <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Ulf Hansson <[email protected]>

Stable-dep-of: 291220451c77 ("mmc: mtk-sd: Fix error handle of probe function")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/mmc/host/mtk-sd.c | 55 ++++++++++++++-------------------------
 1 file changed, 20 insertions(+), 35 deletions(-)

drivers/mmc/host/mtk-sd.c +20 ~~-35~~

index 8b755f1627325..9ebf5aa5d9b18 100644
@@ -2674,20 +2674,18 @@ static int msdc_drv_probe(struct platform_device *pdev)
 	}
 
 	/* Allocate MMC host for this device */
-	mmc = mmc_alloc_host(sizeof(struct msdc_host), &pdev->dev);
+	mmc = devm_mmc_alloc_host(&pdev->dev, sizeof(struct msdc_host));
 	if (!mmc)
 		return -ENOMEM;
 
 	host = mmc_priv(mmc);
 	ret = mmc_of_parse(mmc);
 	if (ret)
-		goto host_free;
+		return ret;
 
 	host->base = devm_platform_ioremap_resource(pdev, 0);
-	if (IS_ERR(host->base)) {
-		ret = PTR_ERR(host->base);
-		goto host_free;
-	}
+	if (IS_ERR(host->base))
+		return PTR_ERR(host->base);
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
 	if (res) {

@@ -2698,18 +2696,16 @@ static int msdc_drv_probe(struct platform_device *pdev)
 
 	ret = mmc_regulator_get_supply(mmc);
 	if (ret)
-		goto host_free;
+		return ret;
 
 	ret = msdc_of_clock_parse(pdev, host);
 	if (ret)
-		goto host_free;
+		return ret;
 
 	host->reset = devm_reset_control_get_optional_exclusive(&pdev->dev,
 								"hrst");
-	if (IS_ERR(host->reset)) {
-		ret = PTR_ERR(host->reset);
-		goto host_free;
-	}
+	if (IS_ERR(host->reset))
+		return PTR_ERR(host->reset);
 
 	/* only eMMC has crypto property */
 	if (!(mmc->caps2 & MMC_CAP2_NO_MMC)) {

@@ -2721,30 +2717,24 @@ static int msdc_drv_probe(struct platform_device *pdev)
 	}
 
 	host->irq = platform_get_irq(pdev, 0);
-	if (host->irq < 0) {
-		ret = host->irq;
-		goto host_free;
-	}
+	if (host->irq < 0)
+		return host->irq;
 
 	host->pinctrl = devm_pinctrl_get(&pdev->dev);
-	if (IS_ERR(host->pinctrl)) {
-		ret = PTR_ERR(host->pinctrl);
-		dev_err(&pdev->dev, "Cannot find pinctrl!\n");
-		goto host_free;
-	}
+	if (IS_ERR(host->pinctrl))
+		return dev_err_probe(&pdev->dev, PTR_ERR(host->pinctrl),
+				     "Cannot find pinctrl");
 
 	host->pins_default = pinctrl_lookup_state(host->pinctrl, "default");
 	if (IS_ERR(host->pins_default)) {
-		ret = PTR_ERR(host->pins_default);
 		dev_err(&pdev->dev, "Cannot find pinctrl default!\n");
-		goto host_free;
+		return PTR_ERR(host->pins_default);
 	}
 
 	host->pins_uhs = pinctrl_lookup_state(host->pinctrl, "state_uhs");
 	if (IS_ERR(host->pins_uhs)) {
-		ret = PTR_ERR(host->pins_uhs);
 		dev_err(&pdev->dev, "Cannot find pinctrl uhs!\n");
-		goto host_free;
+		return PTR_ERR(host->pins_uhs);
 	}
 
 	/* Support for SDIO eint irq ? */

@@ -2833,14 +2823,14 @@ static int msdc_drv_probe(struct platform_device *pdev)
 					     GFP_KERNEL);
 		if (!host->cq_host) {
 			ret = -ENOMEM;
-			goto host_free;
+			goto release_mem;
 		}
 		host->cq_host->caps |= CQHCI_TASK_DESC_SZ_128;
 		host->cq_host->mmio = host->base + 0x800;
 		host->cq_host->ops = &msdc_cmdq_ops;
 		ret = cqhci_init(host->cq_host, mmc, true);
 		if (ret)
-			goto host_free;
+			goto release_mem;
 		mmc->max_segs = 128;
 		/* cqhci 16bit length */
 		/* 0 size, means 65536 so we don't have to -1 here */

@@ -2877,11 +2867,8 @@ static int msdc_drv_probe(struct platform_device *pdev)
 			host->dma.gpd, host->dma.gpd_addr);
 	if (host->dma.bd)
 		dma_free_coherent(&pdev->dev,
-			MAX_BD_NUM * sizeof(struct mt_bdma_desc),
-			host->dma.bd, host->dma.bd_addr);
-host_free:
-	mmc_free_host(mmc);
-
+				  MAX_BD_NUM * sizeof(struct mt_bdma_desc),
+				  host->dma.bd, host->dma.bd_addr);
 	return ret;
 }

@@ -2906,9 +2893,7 @@ static void msdc_drv_remove(struct platform_device *pdev)
 			2 * sizeof(struct mt_gpdma_desc),
 			host->dma.gpd, host->dma.gpd_addr);
 	dma_free_coherent(&pdev->dev, MAX_BD_NUM * sizeof(struct mt_bdma_desc),
-			host->dma.bd, host->dma.bd_addr);
-
-	mmc_free_host(mmc);
+			  host->dma.bd, host->dma.bd_addr);
 }
 
 static void msdc_save_reg(struct msdc_host *host)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #84 lore raw

PATCH 84/356 mmc: mtk-sd: Fix error handle of probe function

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy-ld Lu <[email protected]>

[ Upstream commit 291220451c775a054cedc4fab4578a1419eb6256 ]

In the probe function, it goes to 'release_mem' label and returns after some procedure failure. But if the clocks (partial or all) have been enabled previously, they would not be disabled in msdc_runtime_suspend, since runtime PM is not yet enabled for this case.

That cause mmc related clocks always on during system suspend and block suspend flow. Below log is from a SDCard issue of MT8196 chromebook, it returns -ETIMEOUT while polling clock stable in the msdc_ungate_clock() and probe failed, but the enabled clocks could not be disabled anyway.

[  129.059253] clk_chk_dev_pm_suspend()
[  129.350119] suspend warning: msdcpll is on
[  129.354494] [ck_msdc30_1_sel : enabled, 1, 1, 191999939,   ck_msdcpll_d2]
[  129.362787] [ck_msdcpll_d2   : enabled, 1, 1, 191999939,         msdcpll]
[  129.371041] [ck_msdc30_1_ck  : enabled, 1, 1, 191999939, ck_msdc30_1_sel]
[  129.379295] [msdcpll         : enabled, 1, 1, 383999878,          clk26m]

Add a new 'release_clk' label and reorder the error handle functions to make sure the clocks be disabled after probe failure.

Fixes ffaea6ebfe9c ("mmc: mtk-sd: Use readl_poll_timeout instead of open-coded polling")

Fixes 7a2fa8eed936 ("mmc: mtk-sd: use devm_mmc_alloc_host")

Signed-off-by Andy-ld Lu <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Message-ID: <[email protected]>

Signed-off-by Ulf Hansson <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/mmc/host/mtk-sd.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

drivers/mmc/host/mtk-sd.c +5 -4

index 9ebf5aa5d9b18..a97034388cdff 100644
@@ -2813,7 +2813,7 @@ static int msdc_drv_probe(struct platform_device *pdev)
 	ret = msdc_ungate_clock(host);
 	if (ret) {
 		dev_err(&pdev->dev, "Cannot ungate clocks!\n");
-		goto release_mem;
+		goto release_clk;
 	}
 	msdc_init_hw(host);

@@ -2823,14 +2823,14 @@ static int msdc_drv_probe(struct platform_device *pdev)
 					     GFP_KERNEL);
 		if (!host->cq_host) {
 			ret = -ENOMEM;
-			goto release_mem;
+			goto release;
 		}
 		host->cq_host->caps |= CQHCI_TASK_DESC_SZ_128;
 		host->cq_host->mmio = host->base + 0x800;
 		host->cq_host->ops = &msdc_cmdq_ops;
 		ret = cqhci_init(host->cq_host, mmc, true);
 		if (ret)
-			goto release_mem;
+			goto release;
 		mmc->max_segs = 128;
 		/* cqhci 16bit length */
 		/* 0 size, means 65536 so we don't have to -1 here */

@@ -2857,9 +2857,10 @@ static int msdc_drv_probe(struct platform_device *pdev)
 end:
 	pm_runtime_disable(host->dev);
 release:
-	platform_set_drvdata(pdev, NULL);
 	msdc_deinit_hw(host);
+release_clk:
 	msdc_gate_clock(host);
+	platform_set_drvdata(pdev, NULL);
 release_mem:
 	if (host->dma.gpd)
 		dma_free_coherent(&pdev->dev,

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #85 lore raw

PATCH 85/356 mmc: mtk-sd: fix devm_clk_get_optional usage

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rosen Penev <[email protected]>

[ Upstream commit ed299eda8fbb37cb0e05c7001ab6a6b2627ec087 ]

This already returns NULL when not found. However, it can return EPROBE_DEFER and should thus return here.

Signed-off-by Rosen Penev <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Ulf Hansson <[email protected]>

Stable-dep-of: 2508925fb346 ("mmc: mtk-sd: Fix MMC_CAP2_CRYPTO flag setting")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/mmc/host/mtk-sd.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

drivers/mmc/host/mtk-sd.c +2 -3

index a97034388cdff..c5e96a2c079e5 100644
@@ -2711,9 +2711,8 @@ static int msdc_drv_probe(struct platform_device *pdev)
 	if (!(mmc->caps2 & MMC_CAP2_NO_MMC)) {
 		host->crypto_clk = devm_clk_get_optional(&pdev->dev, "crypto");
 		if (IS_ERR(host->crypto_clk))
-			host->crypto_clk = NULL;
-		else
-			mmc->caps2 |= MMC_CAP2_CRYPTO;
+			return PTR_ERR(host->crypto_clk);
+		mmc->caps2 |= MMC_CAP2_CRYPTO;
 	}
 
 	host->irq = platform_get_irq(pdev, 0);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:34 UTC #86 lore raw

PATCH 86/356 mmc: mtk-sd: Fix MMC_CAP2_CRYPTO flag setting

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy-ld Lu <[email protected]>

[ Upstream commit 2508925fb346661bad9f50b497d7ac7d0b6085d0 ]

Currently, the MMC_CAP2_CRYPTO flag is set by default for eMMC hosts. However, this flag should not be set for hosts that do not support inline encryption.

The 'crypto' clock, as described in the documentation, is used for data encryption and decryption. Therefore, only hosts that are configured with this 'crypto' clock should have the MMC_CAP2_CRYPTO flag set.

Fixes 7b438d0377fb ("mmc: mtk-sd: add Inline Crypto Engine clock control")

Fixes ed299eda8fbb ("mmc: mtk-sd: fix devm_clk_get_optional usage")

Signed-off-by Andy-ld Lu <[email protected]>

Message-ID: <[email protected]>

Signed-off-by Ulf Hansson <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/mmc/host/mtk-sd.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/mmc/host/mtk-sd.c +2 -1

index c5e96a2c079e5..1896bf6746071 100644
@@ -2712,7 +2712,8 @@ static int msdc_drv_probe(struct platform_device *pdev)
 		host->crypto_clk = devm_clk_get_optional(&pdev->dev, "crypto");
 		if (IS_ERR(host->crypto_clk))
 			return PTR_ERR(host->crypto_clk);
-		mmc->caps2 |= MMC_CAP2_CRYPTO;
+		else if (host->crypto_clk)
+			mmc->caps2 |= MMC_CAP2_CRYPTO;
 	}
 
 	host->irq = platform_get_irq(pdev, 0);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #87 lore raw

PATCH 87/356 zram: split memory-tracking and ac-time tracking

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sergey Senozhatsky <[email protected]>

[ Upstream commit a7a0350583ba51d8cde6180bb51d704b89a3b29e ]

ZRAM_MEMORY_TRACKING enables two features:
- per-entry ac-time tracking
- debugfs interface

The latter one is the reason why memory-tracking depends on DEBUG_FS, while the former one is used far beyond debugging these days. Namely ac-time is used for fine grained writeback of idle entries (pages).

Move ac-time tracking under its own config option so that it can be enabled (along with writeback) on systems without DEBUG_FS.

[[email protected]: ifdef fixup, per Dmytro]

Link https://lkml.kernel.org/r/[email protected] open here

Signed-off-by Sergey Senozhatsky <[email protected]>

Cc Minchan Kim <[email protected]>

Cc Dmytro Maluka <[email protected]>

Signed-off-by Andrew Morton <[email protected]>

Stable-dep-of: d37da422edb0 ("zram: clear IDLE flag in mark_idle()")

Signed-off-by Sasha Levin <[email protected]>

---
 Documentation/admin-guide/blockdev/zram.rst |  2 +-
 drivers/block/zram/Kconfig                  | 11 ++++++++-
 drivers/block/zram/zram_drv.c               | 27 ++++++++++-----------
 drivers/block/zram/zram_drv.h               |  2 +-
 4 files changed, 25 insertions(+), 17 deletions(-)

Documentation/admin-guide/blockdev/zram.rst +1 -1

index e4551579cb128..ee2b0030d4168 100644
@@ -328,7 +328,7 @@ as idle::
 From now on, any pages on zram are idle pages. The idle mark
 will be removed until someone requests access of the block.
 IOW, unless there is access request, those pages are still idle pages.
-Additionally, when CONFIG_ZRAM_MEMORY_TRACKING is enabled pages can be
+Additionally, when CONFIG_ZRAM_TRACK_ENTRY_ACTIME is enabled pages can be
 marked as idle based on how long (in seconds) it's been since they were
 last accessed::

drivers/block/zram/Kconfig +10 -1

index 0386b7da02aa3..af201392ed52c 100644
@@ -69,9 +69,18 @@ config ZRAM_WRITEBACK
 
 	 See Documentation/admin-guide/blockdev/zram.rst for more information.
 
+config ZRAM_TRACK_ENTRY_ACTIME
+	bool "Track access time of zram entries"
+	depends on ZRAM
+	help
+	  With this feature zram tracks access time of every stored
+	  entry (page), which can be used for a more fine grained IDLE
+	  pages writeback.
+
 config ZRAM_MEMORY_TRACKING
 	bool "Track zRam block status"
 	depends on ZRAM && DEBUG_FS
+	select ZRAM_TRACK_ENTRY_ACTIME
 	help
 	  With this feature, admin can track the state of allocated blocks
 	  of zRAM. Admin could see the information via

@@ -86,4 +95,4 @@ config ZRAM_MULTI_COMP
 	  This will enable multi-compression streams, so that ZRAM can
 	  re-compress pages using a potentially slower but more effective
 	  compression algorithm. Note, that IDLE page recompression
-	  requires ZRAM_MEMORY_TRACKING.
+	  requires ZRAM_TRACK_ENTRY_ACTIME.

drivers/block/zram/zram_drv.c +13 ~~-14~~

index c29c471b6a182..582df13bfde94 100644
@@ -174,6 +174,14 @@ static inline u32 zram_get_priority(struct zram *zram, u32 index)
 	return prio & ZRAM_COMP_PRIORITY_MASK;
 }
 
+static void zram_accessed(struct zram *zram, u32 index)
+{
+	zram_clear_flag(zram, index, ZRAM_IDLE);
+#ifdef CONFIG_ZRAM_TRACK_ENTRY_ACTIME
+	zram->table[index].ac_time = ktime_get_boottime();
+#endif
+}
+
 static inline void update_used_max(struct zram *zram,
 					const unsigned long pages)
 {

@@ -293,8 +301,9 @@ static void mark_idle(struct zram *zram, ktime_t cutoff)
 		zram_slot_lock(zram, index);
 		if (zram_allocated(zram, index) &&
 				!zram_test_flag(zram, index, ZRAM_UNDER_WB)) {
-#ifdef CONFIG_ZRAM_MEMORY_TRACKING
-			is_idle = !cutoff || ktime_after(cutoff, zram->table[index].ac_time);
+#ifdef CONFIG_ZRAM_TRACK_ENTRY_ACTIME
+			is_idle = !cutoff || ktime_after(cutoff,
+							 zram->table[index].ac_time);
 #endif
 			if (is_idle)
 				zram_set_flag(zram, index, ZRAM_IDLE);

@@ -317,7 +326,7 @@ static ssize_t idle_store(struct device *dev,
 		 */
 		u64 age_sec;
 
-		if (IS_ENABLED(CONFIG_ZRAM_MEMORY_TRACKING) && !kstrtoull(buf, 0, &age_sec))
+		if (IS_ENABLED(CONFIG_ZRAM_TRACK_ENTRY_ACTIME) && !kstrtoull(buf, 0, &age_sec))
 			cutoff_time = ktime_sub(ktime_get_boottime(),
 					ns_to_ktime(age_sec * NSEC_PER_SEC));
 		else

@@ -844,12 +853,6 @@ static void zram_debugfs_destroy(void)
 	debugfs_remove_recursive(zram_debugfs_root);
 }
 
-static void zram_accessed(struct zram *zram, u32 index)
-{
-	zram_clear_flag(zram, index, ZRAM_IDLE);
-	zram->table[index].ac_time = ktime_get_boottime();
-}
-
 static ssize_t read_block_state(struct file *file, char __user *buf,
 				size_t count, loff_t *ppos)
 {

@@ -933,10 +936,6 @@ static void zram_debugfs_unregister(struct zram *zram)
 #else
 static void zram_debugfs_create(void) {};
 static void zram_debugfs_destroy(void) {};
-static void zram_accessed(struct zram *zram, u32 index)
-{
-	zram_clear_flag(zram, index, ZRAM_IDLE);
-};
 static void zram_debugfs_register(struct zram *zram) {};
 static void zram_debugfs_unregister(struct zram *zram) {};
 #endif

@@ -1257,7 +1256,7 @@ static void zram_free_page(struct zram *zram, size_t index)
 {
 	unsigned long handle;
 
-#ifdef CONFIG_ZRAM_MEMORY_TRACKING
+#ifdef CONFIG_ZRAM_TRACK_ENTRY_ACTIME
 	zram->table[index].ac_time = 0;
 #endif
 	if (zram_test_flag(zram, index, ZRAM_IDLE))

drivers/block/zram/zram_drv.h +1 -1

index ca7a15bd48456..35e3221446292 100644
@@ -69,7 +69,7 @@ struct zram_table_entry {
 		unsigned long element;
 	};
 	unsigned long flags;
-#ifdef CONFIG_ZRAM_MEMORY_TRACKING
+#ifdef CONFIG_ZRAM_TRACK_ENTRY_ACTIME
 	ktime_t ac_time;
 #endif
 };

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #88 lore raw

PATCH 88/356 zram: do not mark idle slots that cannot be idle

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sergey Senozhatsky <[email protected]>

[ Upstream commit b967fa1ba72b5da2b6d9bf95f0b13420a59e0701 ]

ZRAM_SAME slots cannot be post-processed (writeback or recompress) so do not mark them ZRAM_IDLE. Same with ZRAM_WB slots, they cannot be ZRAM_IDLE because they are not in zsmalloc pool anymore.

Link https://lkml.kernel.org/r/[email protected] open here

Signed-off-by Sergey Senozhatsky <[email protected]>

Cc Minchan Kim <[email protected]>

Signed-off-by Andrew Morton <[email protected]>

Stable-dep-of: d37da422edb0 ("zram: clear IDLE flag in mark_idle()")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/block/zram/zram_drv.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

drivers/block/zram/zram_drv.c +18 -7

index 582df13bfde94..e05eace18cc47 100644
@@ -297,17 +297,28 @@ static void mark_idle(struct zram *zram, ktime_t cutoff)
 		/*
 		 * Do not mark ZRAM_UNDER_WB slot as ZRAM_IDLE to close race.
 		 * See the comment in writeback_store.
+		 *
+		 * Also do not mark ZRAM_SAME slots as ZRAM_IDLE, because no
+		 * post-processing (recompress, writeback) happens to the
+		 * ZRAM_SAME slot.
+		 *
+		 * And ZRAM_WB slots simply cannot be ZRAM_IDLE.
 		 */
 		zram_slot_lock(zram, index);
-		if (zram_allocated(zram, index) &&
-				!zram_test_flag(zram, index, ZRAM_UNDER_WB)) {
+		if (!zram_allocated(zram, index) ||
+		    zram_test_flag(zram, index, ZRAM_WB) ||
+		    zram_test_flag(zram, index, ZRAM_UNDER_WB) ||
+		    zram_test_flag(zram, index, ZRAM_SAME)) {
+			zram_slot_unlock(zram, index);
+			continue;
+		}
+
 #ifdef CONFIG_ZRAM_TRACK_ENTRY_ACTIME
-			is_idle = !cutoff || ktime_after(cutoff,
-							 zram->table[index].ac_time);
+		is_idle = !cutoff ||
+			ktime_after(cutoff, zram->table[index].ac_time);
 #endif
-			if (is_idle)
-				zram_set_flag(zram, index, ZRAM_IDLE);
-		}
+		if (is_idle)
+			zram_set_flag(zram, index, ZRAM_IDLE);
 		zram_slot_unlock(zram, index);
 	}
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #89 lore raw

PATCH 89/356 zram: clear IDLE flag in mark_idle()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sergey Senozhatsky <[email protected]>

[ Upstream commit d37da422edb0664a2037e6d7d42fe6d339aae78a ]

If entry does not fulfill current mark_idle() parameters, e.g. cutoff time, then we should clear its ZRAM_IDLE from previous mark_idle() invocations.

Consider the following case:
- mark_idle() cutoff time 8h
- mark_idle() cutoff time 4h
- writeback() idle - will writeback entries with cutoff time 8h, while it should only pick entries with cutoff time 4h

The bug was reported by Shin Kawamura.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 755804d16965 ("zram: introduce an aged idle interface")

Signed-off-by Sergey Senozhatsky <[email protected]>

Reported-by Shin Kawamura <[email protected]>

Acked-by Brian Geffon <[email protected]>

Cc Minchan Kim <[email protected]>

Signed-off-by Andrew Morton <[email protected]>

Cc <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/block/zram/zram_drv.c | 2 ++
 1 file changed, 2 insertions(+)

drivers/block/zram/zram_drv.c +2 -0

index e05eace18cc47..27f9ae16a7282 100644
@@ -319,6 +319,8 @@ static void mark_idle(struct zram *zram, ktime_t cutoff)
 #endif
 		if (is_idle)
 			zram_set_flag(zram, index, ZRAM_IDLE);
+		else
+			zram_clear_flag(zram, index, ZRAM_IDLE);
 		zram_slot_unlock(zram, index);
 	}
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #90 lore raw

PATCH 90/356 iommu/arm-smmu: Defer probe of clients after smmu device bound

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pratyush Brahma <[email protected]>

commit 229e6ee43d2a160a1592b83aad620d6027084aad upstream.

Null pointer dereference occurs due to a race between smmu driver probe and client driver probe, when of_dma_configure() for client is called after the iommu_device_register() for smmu driver probe has executed but before the driver_bound() for smmu driver has been called.

Following is how the race occurs:

T1:Smmu device probe T2: Client device probe

really_probe()
arm_smmu_device_probe()
iommu_device_register()

					really_probe()
					platform_dma_configure()
					of_dma_configure()
					of_dma_configure_id()
					of_iommu_configure()
					iommu_probe_device()
					iommu_init_device()
					arm_smmu_probe_device()
					arm_smmu_get_by_fwnode()
						driver_find_device_by_fwnode()
						driver_find_device()
						next_device()
						klist_next()
						    /* null ptr
						       assigned to smmu */
					/* null ptr dereference
					   while smmu->streamid_mask */

driver_bound()
klist_add_tail()

When this null smmu pointer is dereferenced later in arm_smmu_probe_device, the device crashes.

Fix this by deferring the probe of the client device until the smmu device has bound to the arm smmu driver.

Fixes 021bb8420d44 ("iommu/arm-smmu: Wire up generic configuration support")

Co-developed-by Prakash Gupta <[email protected]>

Signed-off-by Prakash Gupta <[email protected]>

Signed-off-by Pratyush Brahma <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

[will: Add comment]

Signed-off-by Will Deacon <[email protected]>

[rm: backport for context conflict prior to 6.8]

Signed-off-by Robin Murphy <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/iommu/arm/arm-smmu/arm-smmu.c |   11 +++++++++++
 1 file changed, 11 insertions(+)

drivers/iommu/arm/arm-smmu/arm-smmu.c +11 -0

@@ -1359,6 +1359,17 @@ static struct iommu_device *arm_smmu_pro
 			goto out_free;
 	} else if (fwspec && fwspec->ops == &arm_smmu_ops) {
 		smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
+
+		/*
+		 * Defer probe if the relevant SMMU instance hasn't finished
+		 * probing yet. This is a fragile hack and we'd ideally
+		 * avoid this race in the core code. Until that's ironed
+		 * out, however, this is the most pragmatic option on the
+		 * table.
+		 */
+		if (!smmu)
+			return ERR_PTR(dev_err_probe(dev, -EPROBE_DEFER,
+						"smmu dev has not bound yet\n"));
 	} else {
 		return ERR_PTR(-ENODEV);
 	}

Greg Kroah-Hartman

author maintainer

2024-12-12 15:37 UTC #91 lore raw

PATCH 91/356 powerpc/vdso: Refactor CFLAGS for CVDSO build

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christophe Leroy <[email protected]>

[ Upstream commit a6b67eb09963af29991625862cbb4f56b85954ed ]

In order to avoid two much duplication when we add new VDSO functionnalities in C like getrandom, refactor common CFLAGS.

Acked-by Michael Ellerman <[email protected]>

Signed-off-by Jason A. Donenfeld <[email protected]>

Stable-dep-of: d677ce521334 ("powerpc/vdso: Drop -mstack-protector-guard flags in 32-bit files with clang")

Signed-off-by Sasha Levin <[email protected]>

---
 arch/powerpc/kernel/vdso/Makefile | 32 +++++++++++++------------------
 1 file changed, 13 insertions(+), 19 deletions(-)

arch/powerpc/kernel/vdso/Makefile +13 ~~-19~~

index 0c7d82c270c37..5c7af93018865 100644
@@ -10,28 +10,11 @@ obj-vdso64 = sigtramp64-64.o gettimeofday-64.o datapage-64.o cacheflush-64.o not
 
 ifneq ($(c-gettimeofday-y),)
   CFLAGS_vgettimeofday-32.o += -include $(c-gettimeofday-y)
-  CFLAGS_vgettimeofday-32.o += $(DISABLE_LATENT_ENTROPY_PLUGIN)
-  CFLAGS_vgettimeofday-32.o += $(call cc-option, -fno-stack-protector)
-  CFLAGS_vgettimeofday-32.o += -DDISABLE_BRANCH_PROFILING
-  CFLAGS_vgettimeofday-32.o += -ffreestanding -fasynchronous-unwind-tables
-  CFLAGS_REMOVE_vgettimeofday-32.o = $(CC_FLAGS_FTRACE)
-  CFLAGS_REMOVE_vgettimeofday-32.o += -mcmodel=medium -mabi=elfv1 -mabi=elfv2 -mcall-aixdesc
-  # This flag is supported by clang for 64-bit but not 32-bit so it will cause
-  # an unused command line flag warning for this file.
-  ifdef CONFIG_CC_IS_CLANG
-  CFLAGS_REMOVE_vgettimeofday-32.o += -fno-stack-clash-protection
-  endif
-  CFLAGS_vgettimeofday-64.o += -include $(c-gettimeofday-y)
-  CFLAGS_vgettimeofday-64.o += $(DISABLE_LATENT_ENTROPY_PLUGIN)
-  CFLAGS_vgettimeofday-64.o += $(call cc-option, -fno-stack-protector)
-  CFLAGS_vgettimeofday-64.o += -DDISABLE_BRANCH_PROFILING
-  CFLAGS_vgettimeofday-64.o += -ffreestanding -fasynchronous-unwind-tables
-  CFLAGS_REMOVE_vgettimeofday-64.o = $(CC_FLAGS_FTRACE)
 # Go prior to 1.16.x assumes r30 is not clobbered by any VDSO code. That used to be true
 # by accident when the VDSO was hand-written asm code, but may not be now that the VDSO is
 # compiler generated. To avoid breaking Go tell GCC not to use r30. Impact on code
 # generation is minimal, it will just use r29 instead.
-  CFLAGS_vgettimeofday-64.o += $(call cc-option, -ffixed-r30)
+  CFLAGS_vgettimeofday-64.o += -include $(c-gettimeofday-y) $(call cc-option, -ffixed-r30)
 endif
 
 # Build rules

@@ -54,6 +37,11 @@ KASAN_SANITIZE := n
 KCSAN_SANITIZE := n
 
 ccflags-y := -fno-common -fno-builtin
+ccflags-y += $(DISABLE_LATENT_ENTROPY_PLUGIN)
+ccflags-y += $(call cc-option, -fno-stack-protector)
+ccflags-y += -DDISABLE_BRANCH_PROFILING
+ccflags-y += -ffreestanding -fasynchronous-unwind-tables
+ccflags-remove-y := $(CC_FLAGS_FTRACE)
 ldflags-y := -Wl,--hash-style=both -nostdlib -shared -z noexecstack $(CLANG_FLAGS)
 ldflags-$(CONFIG_LD_IS_LLD) += $(call cc-option,--ld-path=$(LD),-fuse-ld=lld)
 ldflags-$(CONFIG_LD_ORPHAN_WARN) += -Wl,--orphan-handling=$(CONFIG_LD_ORPHAN_WARN_LEVEL)

@@ -62,6 +50,12 @@ ldflags-$(CONFIG_LD_ORPHAN_WARN) += -Wl,--orphan-handling=$(CONFIG_LD_ORPHAN_WAR
 ldflags-y += $(filter-out $(CC_AUTO_VAR_INIT_ZERO_ENABLER) $(CC_FLAGS_FTRACE) -Wa$(comma)%, $(KBUILD_CFLAGS))
 
 CC32FLAGS := -m32
+CC32FLAGSREMOVE := -mcmodel=medium -mabi=elfv1 -mabi=elfv2 -mcall-aixdesc
+  # This flag is supported by clang for 64-bit but not 32-bit so it will cause
+  # an unused command line flag warning for this file.
+ifdef CONFIG_CC_IS_CLANG
+CC32FLAGSREMOVE += -fno-stack-clash-protection
+endif
 LD32FLAGS := -Wl,-soname=linux-vdso32.so.1
 AS32FLAGS := -D__VDSO32__

@@ -108,7 +102,7 @@ quiet_cmd_vdso32ld_and_check = VDSO32L $@
 quiet_cmd_vdso32as = VDSO32A $@
       cmd_vdso32as = $(VDSOCC) $(a_flags) $(CC32FLAGS) $(AS32FLAGS) -c -o $@ $<
 quiet_cmd_vdso32cc = VDSO32C $@
-      cmd_vdso32cc = $(VDSOCC) $(c_flags) $(CC32FLAGS) -c -o $@ $<
+      cmd_vdso32cc = $(VDSOCC) $(filter-out $(CC32FLAGSREMOVE), $(c_flags)) $(CC32FLAGS) -c -o $@ $<
 
 quiet_cmd_vdso64ld_and_check = VDSO64L $@
       cmd_vdso64ld_and_check = $(VDSOCC) $(ldflags-y) $(LD64FLAGS) -o $@ -Wl,-T$(filter %.lds,$^) $(filter %.o,$^); $(cmd_vdso_check)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #92 lore raw

PATCH 92/356 powerpc/vdso: Drop -mstack-protector-guard flags in 32-bit files with clang

Link https://patch.msgid.link/20241030-powerpc-vdso-drop-stackp-flags-clang-v1-1-d95e7376d29c@kernel.org open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nathan Chancellor <[email protected]>

[ Upstream commit d677ce521334d8f1f327cafc8b1b7854b0833158 ]

Under certain conditions, the 64-bit '-mstack-protector-guard' flags may end up in the 32-bit vDSO flags, resulting in build failures due to the structure of clang's argument parsing of the stack protector options, which validates the arguments of the stack protector guard flags unconditionally in the frontend, choking on the 64-bit values when targeting 32-bit:

clang: error: invalid value 'r13' in 'mstack-protector-guard-reg=', expected one of: r2
clang: error: invalid value 'r13' in 'mstack-protector-guard-reg=', expected one of: r2 make[3]: *** [arch/powerpc/kernel/vdso/Makefile:85: arch/powerpc/kernel/vdso/vgettimeofday-32.o] Error 1 make[3]: *** [arch/powerpc/kernel/vdso/Makefile:87: arch/powerpc/kernel/vdso/vgetrandom-32.o] Error 1

Remove these flags by adding them to the CC32FLAGSREMOVE variable, which already handles situations similar to this. Additionally, reformat and align a comment better for the expanding CONFIG_CC_IS_CLANG block.

Cc [email protected] # v6.1+

Signed-off-by Nathan Chancellor <[email protected]>

Signed-off-by Michael Ellerman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/powerpc/kernel/vdso/Makefile | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

arch/powerpc/kernel/vdso/Makefile +6 -2

index 5c7af93018865..d5defff8472da 100644
@@ -51,10 +51,14 @@ ldflags-y += $(filter-out $(CC_AUTO_VAR_INIT_ZERO_ENABLER) $(CC_FLAGS_FTRACE) -W
 
 CC32FLAGS := -m32
 CC32FLAGSREMOVE := -mcmodel=medium -mabi=elfv1 -mabi=elfv2 -mcall-aixdesc
-  # This flag is supported by clang for 64-bit but not 32-bit so it will cause
-  # an unused command line flag warning for this file.
 ifdef CONFIG_CC_IS_CLANG
+# This flag is supported by clang for 64-bit but not 32-bit so it will cause
+# an unused command line flag warning for this file.
 CC32FLAGSREMOVE += -fno-stack-clash-protection
+# -mstack-protector-guard values from the 64-bit build are not valid for the
+# 32-bit one. clang validates the values passed to these arguments during
+# parsing, even when -fno-stack-protector is passed afterwards.
+CC32FLAGSREMOVE += -mstack-protector-guard%
 endif
 LD32FLAGS := -Wl,-soname=linux-vdso32.so.1
 AS32FLAGS := -D__VDSO32__

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:36 UTC #93 lore raw

PATCH 93/356 ntp: Remove invalid cast in time offset math

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marcelo Dalmas <[email protected]>

commit f5807b0606da7ac7c1b74a386b22134ec7702d05 upstream.

Due to an unsigned cast, adjtimex() returns the wrong offest when using ADJ_MICRO and the offset is negative. In this case a small negative offset returns approximately 4.29 seconds (~ 2^32/1000 milliseconds) due to the unsigned cast of the negative offset.

This cast was added when the kernel internal struct timex was changed to use type long long for the time offset value to address the problem of a 64bit/32bit division on 32bit systems.

The correct cast would have been (s32), which is correct as time_offset can only be in the range of [INT_MIN..INT_MAX] because the shift constant used for calculating it is 32. But that's non-obvious.

Remove the cast and use div_s64() to cure the issue.

[ tglx: Fix white space damage, use div_s64() and amend the change log ]

Fixes ead25417f82e ("timex: use __kernel_timex internally")

Signed-off-by Marcelo Dalmas <[email protected]>

Signed-off-by Thomas Gleixner <[email protected]>

Link https://lore.kernel.org/all/SJ0P101MB03687BF7D5A10FD3C49C51E5F42E2@SJ0P101MB0368.NAMP101.PROD.OUTLOOK.COM open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 kernel/time/ntp.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

kernel/time/ntp.c +1 -1

@@ -796,7 +796,7 @@ int __do_adjtimex(struct __kernel_timex
 		txc->offset = shift_right(time_offset * NTP_INTERVAL_FREQ,
 				  NTP_SCALE_SHIFT);
 		if (!(time_status & STA_NANO))
-			txc->offset = (u32)txc->offset / NSEC_PER_USEC;
+			txc->offset = div_s64(txc->offset, NSEC_PER_USEC);
 	}
 
 	result = time_state;	/* mostly `TIME_OK' */

Greg Kroah-Hartman

author maintainer

2024-12-12 15:37 UTC #94 lore raw

PATCH 94/356 driver core: fw_devlink: Improve logs for cycle detection

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Saravana Kannan <[email protected]>

[ Upstream commit 6e7ad1aebb4fc9fed0217dd50ef6e58a53f17d81 ]

The links in a cycle are not all logged in a consistent manner or not logged at all. Make them consistent by adding a "cycle:" string and log all the link in the cycles (even the child ==> parent dependency) so that it's easier to debug cycle detection code. Also, mark the start and end of a cycle so it's easy to tell when multiple cycles are logged back to back.

Signed-off-by Saravana Kannan <[email protected]>

Tested-by Xu Yang <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Stable-dep-of: bac3b10b78e5 ("driver core: fw_devlink: Stop trying to optimize cycle detection logic")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/base/core.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

drivers/base/core.c +9 -2

index c9fb3243e353e..499904f1ba6b3 100644
@@ -125,7 +125,7 @@ static void __fwnode_link_del(struct fwnode_link *link)
  */
 static void __fwnode_link_cycle(struct fwnode_link *link)
 {
-	pr_debug("%pfwf: Relaxing link with %pfwf\n",
+	pr_debug("%pfwf: cycle: depends on %pfwf\n",
 		 link->consumer, link->supplier);
 	link->flags |= FWLINK_FLAG_CYCLE;
 }

@@ -1959,6 +1959,7 @@ static bool __fw_devlink_relax_cycles(struct device *con,
 
 	/* Termination condition. */
 	if (sup_dev == con) {
+		pr_debug("----- cycle: start -----\n");
 		ret = true;
 		goto out;
 	}

@@ -1990,8 +1991,11 @@ static bool __fw_devlink_relax_cycles(struct device *con,
 	else
 		par_dev = fwnode_get_next_parent_dev(sup_handle);
 
-	if (par_dev && __fw_devlink_relax_cycles(con, par_dev->fwnode))
+	if (par_dev && __fw_devlink_relax_cycles(con, par_dev->fwnode)) {
+		pr_debug("%pfwf: cycle: child of %pfwf\n", sup_handle,
+			 par_dev->fwnode);
 		ret = true;
+	}
 
 	if (!sup_dev)
 		goto out;

@@ -2007,6 +2011,8 @@ static bool __fw_devlink_relax_cycles(struct device *con,
 
 		if (__fw_devlink_relax_cycles(con,
 					      dev_link->supplier->fwnode)) {
+			pr_debug("%pfwf: cycle: depends on %pfwf\n", sup_handle,
+				 dev_link->supplier->fwnode);
 			fw_devlink_relax_link(dev_link);
 			dev_link->flags |= DL_FLAG_CYCLE;
 			ret = true;

@@ -2086,6 +2092,7 @@ static int fw_devlink_create_devlink(struct device *con,
 		if (__fw_devlink_relax_cycles(con, sup_handle)) {
 			__fwnode_link_cycle(link);
 			flags = fw_devlink_get_flags(link->flags);
+			pr_debug("----- cycle: end -----\n");
 			dev_info(con, "Fixed dependency cycle(s) with %pfwf\n",
 				 sup_handle);
 		}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:37 UTC #95 lore raw

PATCH 95/356 driver core: Add FWLINK_FLAG_IGNORE to completely ignore a fwnode link

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Saravana Kannan <[email protected]>

[ Upstream commit b7e1241d8f77ed64404a5e4450f43a319310fc91 ]

A fwnode link between specific supplier-consumer fwnodes can be added multiple times for multiple reasons. If that dependency doesn't exist, deleting the fwnode link once doesn't guarantee that it won't get created again.

So, add FWLINK_FLAG_IGNORE flag to mark a fwnode link as one that needs to be completely ignored. Since a fwnode link's flags is an OR of all the flags passed to all the fwnode_link_add() calls to create that specific fwnode link, the FWLINK_FLAG_IGNORE flag is preserved and can be used to mark a fwnode link as on that need to be completely ignored until it is deleted.

Signed-off-by Saravana Kannan <[email protected]>

Acked-by "Rafael J. Wysocki" <[email protected]>

Reviewed-by Rob Herring <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Stable-dep-of: bac3b10b78e5 ("driver core: fw_devlink: Stop trying to optimize cycle detection logic")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/base/core.c    | 9 ++++++++-
 include/linux/fwnode.h | 2 ++
 2 files changed, 10 insertions(+), 1 deletion(-)

drivers/base/core.c +8 -1

index 499904f1ba6b3..3aaf234dbb088 100644
@@ -1025,7 +1025,8 @@ static struct fwnode_handle *fwnode_links_check_suppliers(
 		return NULL;
 
 	list_for_each_entry(link, &fwnode->suppliers, c_hook)
-		if (!(link->flags & FWLINK_FLAG_CYCLE))
+		if (!(link->flags &
+		      (FWLINK_FLAG_CYCLE | FWLINK_FLAG_IGNORE)))
 			return link->supplier;
 
 	return NULL;

@@ -1976,6 +1977,9 @@ static bool __fw_devlink_relax_cycles(struct device *con,
 	}
 
 	list_for_each_entry(link, &sup_handle->suppliers, c_hook) {
+		if (link->flags & FWLINK_FLAG_IGNORE)
+			continue;
+
 		if (__fw_devlink_relax_cycles(con, link->supplier)) {
 			__fwnode_link_cycle(link);
 			ret = true;

@@ -2054,6 +2058,9 @@ static int fw_devlink_create_devlink(struct device *con,
 	int ret = 0;
 	u32 flags;
 
+	if (link->flags & FWLINK_FLAG_IGNORE)
+		return 0;
+
 	if (con->fwnode == link->consumer)
 		flags = fw_devlink_get_flags(link->flags);
 	else

include/linux/fwnode.h +2 -0

index 5700451b300fb..525cc031596b6 100644
@@ -51,8 +51,10 @@ struct fwnode_handle {
  * fwnode link flags
  *
  * CYCLE:	The fwnode link is part of a cycle. Don't defer probe.
+ * IGNORE:	Completely ignore this link, even during cycle detection.
  */
 #define FWLINK_FLAG_CYCLE			BIT(0)
+#define FWLINK_FLAG_IGNORE			BIT(1)
 
 struct fwnode_link {
 	struct fwnode_handle *supplier;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:37 UTC #96 lore raw

PATCH 96/356 driver core: fw_devlink: Stop trying to optimize cycle detection logic

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Saravana Kannan <[email protected]>

[ Upstream commit bac3b10b78e54b7da3cede397258f75a2180609b ]

In attempting to optimize fw_devlink runtime, I introduced numerous cycle detection bugs by foregoing cycle detection logic under specific conditions. Each fix has further narrowed the conditions for optimization.

It's time to give up on these optimization attempts and just run the cycle detection logic every time fw_devlink tries to create a device link.

The specific bug report that triggered this fix involved a supplier fwnode that never gets a device created for it. Instead, the supplier fwnode is represented by the device that corresponds to an ancestor fwnode.

In this case, fw_devlink didn't do any cycle detection because the cycle detection logic is only run when a device link is created between the devices that correspond to the actual consumer and supplier fwnodes.

With this change, fw_devlink will run cycle detection logic even when creating SYNC_STATE_ONLY proxy device links from a device that is an ancestor of a consumer fwnode.

Reported-by Tomi Valkeinen <[email protected]>

Closes https://lore.kernel.org/all/[email protected]/ open here

Fixes 6442d79d880c ("driver core: fw_devlink: Improve detection of overlapping cycles")

Cc stable <[email protected]>

Tested-by Tomi Valkeinen <[email protected]>

Signed-off-by Saravana Kannan <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/base/core.c | 55 ++++++++++++++++++++-------------------------
 1 file changed, 24 insertions(+), 31 deletions(-)

drivers/base/core.c +24 ~~-31~~

index 3aaf234dbb088..18a73e4921026 100644
@@ -1936,10 +1936,10 @@ static bool fwnode_ancestor_init_without_drv(struct fwnode_handle *fwnode)
  *
  * Return true if one or more cycles were found. Otherwise, return false.
  */
-static bool __fw_devlink_relax_cycles(struct device *con,
+static bool __fw_devlink_relax_cycles(struct fwnode_handle *con_handle,
 				 struct fwnode_handle *sup_handle)
 {
-	struct device *sup_dev = NULL, *par_dev = NULL;
+	struct device *sup_dev = NULL, *par_dev = NULL, *con_dev = NULL;
 	struct fwnode_link *link;
 	struct device_link *dev_link;
 	bool ret = false;

@@ -1956,22 +1956,22 @@ static bool __fw_devlink_relax_cycles(struct device *con,
 
 	sup_handle->flags |= FWNODE_FLAG_VISITED;
 
-	sup_dev = get_dev_from_fwnode(sup_handle);
-
 	/* Termination condition. */
-	if (sup_dev == con) {
+	if (sup_handle == con_handle) {
 		pr_debug("----- cycle: start -----\n");
 		ret = true;
 		goto out;
 	}
 
+	sup_dev = get_dev_from_fwnode(sup_handle);
+	con_dev = get_dev_from_fwnode(con_handle);
 	/*
 	 * If sup_dev is bound to a driver and @con hasn't started binding to a
 	 * driver, sup_dev can't be a consumer of @con. So, no need to check
 	 * further.
 	 */
 	if (sup_dev && sup_dev->links.status ==  DL_DEV_DRIVER_BOUND &&
-	    con->links.status == DL_DEV_NO_DRIVER) {
+	    con_dev && con_dev->links.status == DL_DEV_NO_DRIVER) {
 		ret = false;
 		goto out;
 	}

@@ -1980,7 +1980,7 @@ static bool __fw_devlink_relax_cycles(struct device *con,
 		if (link->flags & FWLINK_FLAG_IGNORE)
 			continue;
 
-		if (__fw_devlink_relax_cycles(con, link->supplier)) {
+		if (__fw_devlink_relax_cycles(con_handle, link->supplier)) {
 			__fwnode_link_cycle(link);
 			ret = true;
 		}

@@ -1995,7 +1995,7 @@ static bool __fw_devlink_relax_cycles(struct device *con,
 	else
 		par_dev = fwnode_get_next_parent_dev(sup_handle);
 
-	if (par_dev && __fw_devlink_relax_cycles(con, par_dev->fwnode)) {
+	if (par_dev && __fw_devlink_relax_cycles(con_handle, par_dev->fwnode)) {
 		pr_debug("%pfwf: cycle: child of %pfwf\n", sup_handle,
 			 par_dev->fwnode);
 		ret = true;

@@ -2013,7 +2013,7 @@ static bool __fw_devlink_relax_cycles(struct device *con,
 		    !(dev_link->flags & DL_FLAG_CYCLE))
 			continue;
 
-		if (__fw_devlink_relax_cycles(con,
+		if (__fw_devlink_relax_cycles(con_handle,
 					      dev_link->supplier->fwnode)) {
 			pr_debug("%pfwf: cycle: depends on %pfwf\n", sup_handle,
 				 dev_link->supplier->fwnode);

@@ -2061,11 +2061,6 @@ static int fw_devlink_create_devlink(struct device *con,
 	if (link->flags & FWLINK_FLAG_IGNORE)
 		return 0;
 
-	if (con->fwnode == link->consumer)
-		flags = fw_devlink_get_flags(link->flags);
-	else
-		flags = FW_DEVLINK_FLAGS_PERMISSIVE;
-
 	/*
 	 * In some cases, a device P might also be a supplier to its child node
 	 * C. However, this would defer the probe of C until the probe of P

@@ -2086,25 +2081,23 @@ static int fw_devlink_create_devlink(struct device *con,
 		return -EINVAL;
 
 	/*
-	 * SYNC_STATE_ONLY device links don't block probing and supports cycles.
-	 * So, one might expect that cycle detection isn't necessary for them.
-	 * However, if the device link was marked as SYNC_STATE_ONLY because
-	 * it's part of a cycle, then we still need to do cycle detection. This
-	 * is because the consumer and supplier might be part of multiple cycles
-	 * and we need to detect all those cycles.
+	 * Don't try to optimize by not calling the cycle detection logic under
+	 * certain conditions. There's always some corner case that won't get
+	 * detected.
 	 */
-	if (!device_link_flag_is_sync_state_only(flags) ||
-	    flags & DL_FLAG_CYCLE) {
-		device_links_write_lock();
-		if (__fw_devlink_relax_cycles(con, sup_handle)) {
-			__fwnode_link_cycle(link);
-			flags = fw_devlink_get_flags(link->flags);
-			pr_debug("----- cycle: end -----\n");
-			dev_info(con, "Fixed dependency cycle(s) with %pfwf\n",
-				 sup_handle);
-		}
-		device_links_write_unlock();
+	device_links_write_lock();
+	if (__fw_devlink_relax_cycles(link->consumer, sup_handle)) {
+		__fwnode_link_cycle(link);
+		pr_debug("----- cycle: end -----\n");
+		pr_info("%pfwf: Fixed dependency cycle(s) with %pfwf\n",
+			link->consumer, sup_handle);
 	}
+	device_links_write_unlock();
+
+	if (con->fwnode == link->consumer)
+		flags = fw_devlink_get_flags(link->flags);
+	else
+		flags = FW_DEVLINK_FLAGS_PERMISSIVE;
 
 	if (sup_handle->flags & FWNODE_FLAG_NOT_DEVICE)
 		sup_dev = fwnode_get_next_parent_dev(sup_handle);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:37 UTC #97 lore raw

PATCH 97/356 f2fs: fix to drop all discards after creating snapshot on lvm device

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chao Yu <[email protected]>

[ Upstream commit bc8aeb04fd80cb8cfae3058445c84410fd0beb5e ]

Piergiorgio reported a bug in bugzilla as below:

------------[ cut here ]------------
WARNING: CPU: 2 PID: 969 at fs/f2fs/segment.c:1330
RIP: 0010:__submit_discard_cmd+0x27d/0x400 [f2fs]
Call Trace:
 __issue_discard_cmd+0x1ca/0x350 [f2fs]
 issue_discard_thread+0x191/0x480 [f2fs]
 kthread+0xcf/0x100
 ret_from_fork+0x31/0x50
 ret_from_fork_asm+0x1a/0x30

w/ below testcase, it can reproduce this bug quickly:
- pvcreate /dev/vdb
- vgcreate myvg1 /dev/vdb
- lvcreate -L 1024m -n mylv1 myvg1
- mount /dev/myvg1/mylv1 /mnt/f2fs
- dd if=/dev/zero of=/mnt/f2fs/file bs=1M count=20
- sync
- rm /mnt/f2fs/file
- sync
- lvcreate -L 1024m -s -n mylv1-snapshot /dev/myvg1/mylv1
- umount /mnt/f2fs

The root cause is: it will update discard_max_bytes of mounted lvm device to zero after creating snapshot on this lvm device, then, __submit_discard_cmd() will pass parameter @nr_sects w/ zero value to __blkdev_issue_discard(), it returns a NULL bio pointer, result in panic.

This patch changes as below for fixing:
1. Let's drop all remained discards in f2fs_unfreeze() if snapshot of lvm device is created.
2. Checking discard_max_bytes before submitting discard during __submit_discard_cmd().

Fixes 35ec7d574884 ("f2fs: split discard command in prior to block layer")

Reported-by Piergiorgio Sartor <[email protected]>

Closes https://bugzilla.kernel.org/show_bug.cgi?id=219484

Signed-off-by Chao Yu <[email protected]>

Signed-off-by Jaegeuk Kim <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 fs/f2fs/segment.c |  9 +++++++++
 fs/f2fs/super.c   | 12 ++++++++++++
 2 files changed, 21 insertions(+)

fs/f2fs/segment.c +9 -0

index 670104628ddbe..156d92b945258 100644
@@ -1282,6 +1282,15 @@ static int __submit_discard_cmd(struct f2fs_sb_info *sbi,
 	}
 #endif
 
+	/*
+	 * stop issuing discard for any of below cases:
+	 * 1. device is conventional zone, but it doesn't support discard.
+	 * 2. device is regulare device, after snapshot it doesn't support
+	 * discard.
+	 */
+	if (!bdev_max_discard_sectors(bdev))
+		return -EOPNOTSUPP;
+
 	trace_f2fs_issue_discard(bdev, dc->di.start, dc->di.len);
 
 	lstart = dc->di.lstart;

fs/f2fs/super.c +12 -0

index f05d0e43db9e2..b72fa103b9632 100644
@@ -1735,6 +1735,18 @@ static int f2fs_freeze(struct super_block *sb)
 
 static int f2fs_unfreeze(struct super_block *sb)
 {
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+
+	/*
+	 * It will update discard_max_bytes of mounted lvm device to zero
+	 * after creating snapshot on this lvm device, let's drop all
+	 * remained discards.
+	 * We don't need to disable real-time discard because discard_max_bytes
+	 * will recover after removal of snapshot.
+	 */
+	if (test_opt(sbi, DISCARD) && !f2fs_hw_support_discard(sbi))
+		f2fs_issue_discard_timeout(sbi);
+
 	clear_sbi_flag(F2FS_SB(sb), SBI_IS_FREEZING);
 	return 0;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:37 UTC #98 lore raw

PATCH 98/356 i3c: master: add enable(disable) hot join in sys entry

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Frank Li <[email protected]>

[ Upstream commit 317bacf960a4879af22d12175f47d284930b3273 ]

Add hotjoin entry in sys file system allow user enable/disable hotjoin feature.

Add (*enable(disable)_hotjoin)() to i3c_master_controller_ops. Add api i3c_master_enable(disable)_hotjoin();

Signed-off-by Frank Li <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Alexandre Belloni <[email protected]>

Stable-dep-of: 25bc99be5fe5 ("i3c: master: svc: Modify enabled_events bit 7:0 to act as IBI enable counter")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/i3c/master.c       | 83 ++++++++++++++++++++++++++++++++++++++
 include/linux/i3c/master.h |  5 +++
 2 files changed, 88 insertions(+)

drivers/i3c/master.c +83 -0

index 70d120dfb0908..bbd5dc89be229 100644
@@ -526,6 +526,88 @@ static ssize_t i2c_scl_frequency_show(struct device *dev,
 }
 static DEVICE_ATTR_RO(i2c_scl_frequency);
 
+static int i3c_set_hotjoin(struct i3c_master_controller *master, bool enable)
+{
+	int ret;
+
+	if (!master || !master->ops)
+		return -EINVAL;
+
+	if (!master->ops->enable_hotjoin || !master->ops->disable_hotjoin)
+		return -EINVAL;
+
+	i3c_bus_normaluse_lock(&master->bus);
+
+	if (enable)
+		ret = master->ops->enable_hotjoin(master);
+	else
+		ret = master->ops->disable_hotjoin(master);
+
+	master->hotjoin = enable;
+
+	i3c_bus_normaluse_unlock(&master->bus);
+
+	return ret;
+}
+
+static ssize_t hotjoin_store(struct device *dev, struct device_attribute *attr,
+			     const char *buf, size_t count)
+{
+	struct i3c_bus *i3cbus = dev_to_i3cbus(dev);
+	int ret;
+	bool res;
+
+	if (!i3cbus->cur_master)
+		return -EINVAL;
+
+	if (kstrtobool(buf, &res))
+		return -EINVAL;
+
+	ret = i3c_set_hotjoin(i3cbus->cur_master->common.master, res);
+	if (ret)
+		return ret;
+
+	return count;
+}
+
+/*
+ * i3c_master_enable_hotjoin - Enable hotjoin
+ * @master: I3C master object
+ *
+ * Return: a 0 in case of success, an negative error code otherwise.
+ */
+int i3c_master_enable_hotjoin(struct i3c_master_controller *master)
+{
+	return i3c_set_hotjoin(master, true);
+}
+EXPORT_SYMBOL_GPL(i3c_master_enable_hotjoin);
+
+/*
+ * i3c_master_disable_hotjoin - Disable hotjoin
+ * @master: I3C master object
+ *
+ * Return: a 0 in case of success, an negative error code otherwise.
+ */
+int i3c_master_disable_hotjoin(struct i3c_master_controller *master)
+{
+	return i3c_set_hotjoin(master, false);
+}
+EXPORT_SYMBOL_GPL(i3c_master_disable_hotjoin);
+
+static ssize_t hotjoin_show(struct device *dev, struct device_attribute *da, char *buf)
+{
+	struct i3c_bus *i3cbus = dev_to_i3cbus(dev);
+	ssize_t ret;
+
+	i3c_bus_normaluse_lock(i3cbus);
+	ret = sysfs_emit(buf, "%d\n", i3cbus->cur_master->common.master->hotjoin);
+	i3c_bus_normaluse_unlock(i3cbus);
+
+	return ret;
+}
+
+static DEVICE_ATTR_RW(hotjoin);
+
 static struct attribute *i3c_masterdev_attrs[] = {
 	&dev_attr_mode.attr,
 	&dev_attr_current_master.attr,

@@ -536,6 +618,7 @@ static struct attribute *i3c_masterdev_attrs[] = {
 	&dev_attr_pid.attr,
 	&dev_attr_dynamic_address.attr,
 	&dev_attr_hdrcap.attr,
+	&dev_attr_hotjoin.attr,
 	NULL,
 };
 ATTRIBUTE_GROUPS(i3c_masterdev);

include/linux/i3c/master.h +5 -0

index 0b52da4f23467..65b8965968af2 100644
@@ -452,6 +452,8 @@ struct i3c_master_controller_ops {
 	int (*disable_ibi)(struct i3c_dev_desc *dev);
 	void (*recycle_ibi_slot)(struct i3c_dev_desc *dev,
 				 struct i3c_ibi_slot *slot);
+	int (*enable_hotjoin)(struct i3c_master_controller *master);
+	int (*disable_hotjoin)(struct i3c_master_controller *master);
 };
 
 /**

@@ -487,6 +489,7 @@ struct i3c_master_controller {
 	const struct i3c_master_controller_ops *ops;
 	unsigned int secondary : 1;
 	unsigned int init_done : 1;
+	unsigned int hotjoin: 1;
 	struct {
 		struct list_head i3c;
 		struct list_head i2c;

@@ -543,6 +546,8 @@ int i3c_master_register(struct i3c_master_controller *master,
 			const struct i3c_master_controller_ops *ops,
 			bool secondary);
 void i3c_master_unregister(struct i3c_master_controller *master);
+int i3c_master_enable_hotjoin(struct i3c_master_controller *master);
+int i3c_master_disable_hotjoin(struct i3c_master_controller *master);
 
 /**
  * i3c_dev_get_master_data() - get master private data attached to an I3C

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:37 UTC #99 lore raw

PATCH 99/356 i3c: master: svc: add hot join support

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Frank Li <[email protected]>

[ Upstream commit 05b26c31a4859af9e75b7de77458e99358364fe1 ]

Add hot join support for svc master controller. Disable hot join by default.
User can use sysfs entry to enable hot join.

Signed-off-by Frank Li <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Alexandre Belloni <[email protected]>

Stable-dep-of: 25bc99be5fe5 ("i3c: master: svc: Modify enabled_events bit 7:0 to act as IBI enable counter")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/i3c/master/svc-i3c-master.c | 61 +++++++++++++++++++++++++++--
 1 file changed, 57 insertions(+), 4 deletions(-)

drivers/i3c/master/svc-i3c-master.c +57 -4

index dca266d9dd122..139b51a575366 100644
@@ -128,6 +128,9 @@
 /* This parameter depends on the implementation and may be tuned */
 #define SVC_I3C_FIFO_SIZE 16
 
+#define SVC_I3C_EVENT_IBI	BIT(0)
+#define SVC_I3C_EVENT_HOTJOIN	BIT(1)
+
 struct svc_i3c_cmd {
 	u8 addr;
 	bool rnw;

@@ -178,6 +181,7 @@ struct svc_i3c_regs_save {
  * @ibi.tbq_slot: To be queued IBI slot
  * @ibi.lock: IBI lock
  * @lock: Transfer lock, protect between IBI work thread and callbacks from master
+ * @enabled_events: Bit masks for enable events (IBI, HotJoin).
  */
 struct svc_i3c_master {
 	struct i3c_master_controller base;

@@ -207,6 +211,7 @@ struct svc_i3c_master {
 		spinlock_t lock;
 	} ibi;
 	struct mutex lock;
+	int enabled_events;
 };
 
 /**

@@ -221,6 +226,11 @@ struct svc_i3c_i2c_dev_data {
 	struct i3c_generic_ibi_pool *ibi_pool;
 };
 
+static inline bool is_events_enabled(struct svc_i3c_master *master, u32 mask)
+{
+	return !!(master->enabled_events & mask);
+}
+
 static bool svc_i3c_master_error(struct svc_i3c_master *master)
 {
 	u32 mstatus, merrwarn;

@@ -440,13 +450,16 @@ static void svc_i3c_master_ibi_work(struct work_struct *work)
 	switch (ibitype) {
 	case SVC_I3C_MSTATUS_IBITYPE_IBI:
 		dev = svc_i3c_master_dev_from_addr(master, ibiaddr);
-		if (!dev)
+		if (!dev || !is_events_enabled(master, SVC_I3C_EVENT_IBI))
 			svc_i3c_master_nack_ibi(master);
 		else
 			svc_i3c_master_handle_ibi(master, dev);
 		break;
 	case SVC_I3C_MSTATUS_IBITYPE_HOT_JOIN:
-		svc_i3c_master_ack_ibi(master, false);
+		if (is_events_enabled(master, SVC_I3C_EVENT_HOTJOIN))
+			svc_i3c_master_ack_ibi(master, false);
+		else
+			svc_i3c_master_nack_ibi(master);
 		break;
 	case SVC_I3C_MSTATUS_IBITYPE_MASTER_REQUEST:
 		svc_i3c_master_nack_ibi(master);

@@ -483,7 +496,9 @@ static void svc_i3c_master_ibi_work(struct work_struct *work)
 		svc_i3c_master_emit_stop(master);
 		break;
 	case SVC_I3C_MSTATUS_IBITYPE_HOT_JOIN:
-		queue_work(master->base.wq, &master->hj_work);
+		svc_i3c_master_emit_stop(master);
+		if (is_events_enabled(master, SVC_I3C_EVENT_HOTJOIN))
+			queue_work(master->base.wq, &master->hj_work);
 		break;
 	case SVC_I3C_MSTATUS_IBITYPE_MASTER_REQUEST:
 	default:

@@ -1520,6 +1535,7 @@ static int svc_i3c_master_enable_ibi(struct i3c_dev_desc *dev)
 		return ret;
 	}
 
+	master->enabled_events |= SVC_I3C_EVENT_IBI;
 	svc_i3c_master_enable_interrupts(master, SVC_I3C_MINT_SLVSTART);
 
 	return i3c_master_enec_locked(m, dev->info.dyn_addr, I3C_CCC_EVENT_SIR);

@@ -1531,7 +1547,9 @@ static int svc_i3c_master_disable_ibi(struct i3c_dev_desc *dev)
 	struct svc_i3c_master *master = to_svc_i3c_master(m);
 	int ret;
 
-	svc_i3c_master_disable_interrupts(master);
+	master->enabled_events &= ~SVC_I3C_EVENT_IBI;
+	if (!master->enabled_events)
+		svc_i3c_master_disable_interrupts(master);
 
 	ret = i3c_master_disec_locked(m, dev->info.dyn_addr, I3C_CCC_EVENT_SIR);

@@ -1541,6 +1559,39 @@ static int svc_i3c_master_disable_ibi(struct i3c_dev_desc *dev)
 	return ret;
 }
 
+static int svc_i3c_master_enable_hotjoin(struct i3c_master_controller *m)
+{
+	struct svc_i3c_master *master = to_svc_i3c_master(m);
+	int ret;
+
+	ret = pm_runtime_resume_and_get(master->dev);
+	if (ret < 0) {
+		dev_err(master->dev, "<%s> Cannot get runtime PM.\n", __func__);
+		return ret;
+	}
+
+	master->enabled_events |= SVC_I3C_EVENT_HOTJOIN;
+
+	svc_i3c_master_enable_interrupts(master, SVC_I3C_MINT_SLVSTART);
+
+	return 0;
+}
+
+static int svc_i3c_master_disable_hotjoin(struct i3c_master_controller *m)
+{
+	struct svc_i3c_master *master = to_svc_i3c_master(m);
+
+	master->enabled_events &= ~SVC_I3C_EVENT_HOTJOIN;
+
+	if (!master->enabled_events)
+		svc_i3c_master_disable_interrupts(master);
+
+	pm_runtime_mark_last_busy(master->dev);
+	pm_runtime_put_autosuspend(master->dev);
+
+	return 0;
+}
+
 static void svc_i3c_master_recycle_ibi_slot(struct i3c_dev_desc *dev,
 					    struct i3c_ibi_slot *slot)
 {

@@ -1567,6 +1618,8 @@ static const struct i3c_master_controller_ops svc_i3c_master_ops = {
 	.recycle_ibi_slot = svc_i3c_master_recycle_ibi_slot,
 	.enable_ibi = svc_i3c_master_enable_ibi,
 	.disable_ibi = svc_i3c_master_disable_ibi,
+	.enable_hotjoin = svc_i3c_master_enable_hotjoin,
+	.disable_hotjoin = svc_i3c_master_disable_hotjoin,
 };
 
 static int svc_i3c_master_prepare_clks(struct svc_i3c_master *master)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:37 UTC #100 lore raw

PATCH 100/356 i3c: master: fix kernel-doc check warning

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Frank Li <[email protected]>

[ Upstream commit 34d946b723b53488ab39d8ac540ddf9db255317a ]

Fix warning found by
'scripts/kernel-doc -v -none include/linux/i3c/master.h'

include/linux/i3c/master.h:457: warning: Function parameter or member 'enable_hotjoin' not described in 'i3c_master_controller_ops' include/linux/i3c/master.h:457: warning: Function parameter or member 'disable_hotjoin' not described in 'i3c_master_controller_ops' include/linux/i3c/master.h:499: warning: Function parameter or member 'hotjoin' not described in 'i3c_master_controller'

Signed-off-by Frank Li <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Alexandre Belloni <[email protected]>

Stable-dep-of: 25bc99be5fe5 ("i3c: master: svc: Modify enabled_events bit 7:0 to act as IBI enable counter")

Signed-off-by Sasha Levin <[email protected]>

---
 include/linux/i3c/master.h | 3 +++
 1 file changed, 3 insertions(+)

include/linux/i3c/master.h +3 -0

index 65b8965968af2..1cbf0baca65fe 100644
@@ -426,6 +426,8 @@ struct i3c_bus {
  *		      for a future IBI
  *		      This method is mandatory only if ->request_ibi is not
  *		      NULL.
+ * @enable_hotjoin: enable hot join event detect.
+ * @disable_hotjoin: disable hot join event detect.
  */
 struct i3c_master_controller_ops {
 	int (*bus_init)(struct i3c_master_controller *master);

@@ -467,6 +469,7 @@ struct i3c_master_controller_ops {
  * @ops: master operations. See &struct i3c_master_controller_ops
  * @secondary: true if the master is a secondary master
  * @init_done: true when the bus initialization is done
+ * @hotjoin: true if the master support hotjoin
  * @boardinfo.i3c: list of I3C  boardinfo objects
  * @boardinfo.i2c: list of I2C boardinfo objects
  * @boardinfo: board-level information attached to devices connected on the bus

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #101 lore raw

PATCH 101/356 i3c: master: support to adjust first broadcast address speed

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Carlos Song <[email protected]>

[ Upstream commit aef79e189ba2b32f78bd35daf2c0b41f3868a321 ]

According to I3C spec 6.2 Timing Specification, the Open Drain High Period of SCL Clock timing for first broadcast address should be adjusted to 200ns at least. I3C device working as i2c device will see the broadcast to close its Spike Filter then change to work at I3C mode. After that I3C open drain SCL high level should be adjusted back.

Signed-off-by Carlos Song <[email protected]>

Reviewed-by Frank Li <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Alexandre Belloni <[email protected]>

Stable-dep-of: 25bc99be5fe5 ("i3c: master: svc: Modify enabled_events bit 7:0 to act as IBI enable counter")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/i3c/master.c       | 12 ++++++++++++
 include/linux/i3c/master.h | 16 ++++++++++++++++
 2 files changed, 28 insertions(+)

drivers/i3c/master.c +12 -0

index bbd5dc89be229..78171a754a3f8 100644
@@ -1827,6 +1827,12 @@ static int i3c_master_bus_init(struct i3c_master_controller *master)
 		goto err_bus_cleanup;
 	}
 
+	if (master->ops->set_speed) {
+		ret = master->ops->set_speed(master, I3C_OPEN_DRAIN_SLOW_SPEED);
+		if (ret)
+			goto err_bus_cleanup;
+	}
+
 	/*
 	 * Reset all dynamic address that may have been assigned before
 	 * (assigned by the bootloader for example).

@@ -1835,6 +1841,12 @@ static int i3c_master_bus_init(struct i3c_master_controller *master)
 	if (ret && ret != I3C_ERROR_M2)
 		goto err_bus_cleanup;
 
+	if (master->ops->set_speed) {
+		master->ops->set_speed(master, I3C_OPEN_DRAIN_NORMAL_SPEED);
+		if (ret)
+			goto err_bus_cleanup;
+	}
+
 	/* Disable all slave events before starting DAA. */
 	ret = i3c_master_disec_locked(master, I3C_BROADCAST_ADDR,
 				      I3C_CCC_EVENT_SIR | I3C_CCC_EVENT_MR |

include/linux/i3c/master.h +16 -0

index 1cbf0baca65fe..f0aa4c549aa7f 100644
@@ -269,6 +269,20 @@ enum i3c_bus_mode {
 	I3C_BUS_MODE_MIXED_SLOW,
 };
 
+/**
+ * enum i3c_open_drain_speed - I3C open-drain speed
+ * @I3C_OPEN_DRAIN_SLOW_SPEED: Slow open-drain speed for sending the first
+ *				broadcast address. The first broadcast address at this speed
+ *				will be visible to all devices on the I3C bus. I3C devices
+ *				working in I2C mode will turn off their spike filter when
+ *				switching into I3C mode.
+ * @I3C_OPEN_DRAIN_NORMAL_SPEED: Normal open-drain speed in I3C bus mode.
+ */
+enum i3c_open_drain_speed {
+	I3C_OPEN_DRAIN_SLOW_SPEED,
+	I3C_OPEN_DRAIN_NORMAL_SPEED,
+};
+
 /**
  * enum i3c_addr_slot_status - I3C address slot status
  * @I3C_ADDR_SLOT_FREE: address is free

@@ -428,6 +442,7 @@ struct i3c_bus {
  *		      NULL.
  * @enable_hotjoin: enable hot join event detect.
  * @disable_hotjoin: disable hot join event detect.
+ * @set_speed: adjust I3C open drain mode timing.
  */
 struct i3c_master_controller_ops {
 	int (*bus_init)(struct i3c_master_controller *master);

@@ -456,6 +471,7 @@ struct i3c_master_controller_ops {
 				 struct i3c_ibi_slot *slot);
 	int (*enable_hotjoin)(struct i3c_master_controller *master);
 	int (*disable_hotjoin)(struct i3c_master_controller *master);
+	int (*set_speed)(struct i3c_master_controller *master, enum i3c_open_drain_speed speed);
 };
 
 /**

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #102 lore raw

PATCH 102/356 i3c: master: svc: use slow speed for first broadcast address

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Carlos Song <[email protected]>

[ Upstream commit 20ade67bb1645f5ce8f37fa79ddfebbc5b5b24ef ]

I3C controller should support adjusting open drain timing for the first broadcast address to make I3C device working as a i2c device can see slow broadcast address to close its Spike Filter to change working at i3c mode.

Signed-off-by Carlos Song <[email protected]>

Reviewed-by Frank Li <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Alexandre Belloni <[email protected]>

Stable-dep-of: 25bc99be5fe5 ("i3c: master: svc: Modify enabled_events bit 7:0 to act as IBI enable counter")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/i3c/master/svc-i3c-master.c | 52 +++++++++++++++++++++++++++++
 1 file changed, 52 insertions(+)

drivers/i3c/master/svc-i3c-master.c +52 -0

index 139b51a575366..97d03d755a61f 100644
@@ -182,6 +182,7 @@ struct svc_i3c_regs_save {
  * @ibi.lock: IBI lock
  * @lock: Transfer lock, protect between IBI work thread and callbacks from master
  * @enabled_events: Bit masks for enable events (IBI, HotJoin).
+ * @mctrl_config: Configuration value in SVC_I3C_MCTRL for setting speed back.
  */
 struct svc_i3c_master {
 	struct i3c_master_controller base;

@@ -212,6 +213,7 @@ struct svc_i3c_master {
 	} ibi;
 	struct mutex lock;
 	int enabled_events;
+	u32 mctrl_config;
 };
 
 /**

@@ -529,6 +531,54 @@ static irqreturn_t svc_i3c_master_irq_handler(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
+static int svc_i3c_master_set_speed(struct i3c_master_controller *m,
+				     enum i3c_open_drain_speed speed)
+{
+	struct svc_i3c_master *master = to_svc_i3c_master(m);
+	struct i3c_bus *bus = i3c_master_get_bus(&master->base);
+	u32 ppbaud, odbaud, odhpp, mconfig;
+	unsigned long fclk_rate;
+	int ret;
+
+	ret = pm_runtime_resume_and_get(master->dev);
+	if (ret < 0) {
+		dev_err(master->dev, "<%s> Cannot get runtime PM.\n", __func__);
+		return ret;
+	}
+
+	switch (speed) {
+	case I3C_OPEN_DRAIN_SLOW_SPEED:
+		fclk_rate = clk_get_rate(master->fclk);
+		if (!fclk_rate) {
+			ret = -EINVAL;
+			goto rpm_out;
+		}
+		/*
+		 * Set 50% duty-cycle I2C speed to I3C OPEN-DRAIN mode, so the first
+		 * broadcast address is visible to all I2C/I3C devices on the I3C bus.
+		 * I3C device working as a I2C device will turn off its 50ns Spike
+		 * Filter to change to I3C mode.
+		 */
+		mconfig = master->mctrl_config;
+		ppbaud = FIELD_GET(GENMASK(11, 8), mconfig);
+		odhpp = 0;
+		odbaud = DIV_ROUND_UP(fclk_rate, bus->scl_rate.i2c * (2 + 2 * ppbaud)) - 1;
+		mconfig &= ~GENMASK(24, 16);
+		mconfig |= SVC_I3C_MCONFIG_ODBAUD(odbaud) | SVC_I3C_MCONFIG_ODHPP(odhpp);
+		writel(mconfig, master->regs + SVC_I3C_MCONFIG);
+		break;
+	case I3C_OPEN_DRAIN_NORMAL_SPEED:
+		writel(master->mctrl_config, master->regs + SVC_I3C_MCONFIG);
+		break;
+	}
+
+rpm_out:
+	pm_runtime_mark_last_busy(master->dev);
+	pm_runtime_put_autosuspend(master->dev);
+
+	return ret;
+}
+
 static int svc_i3c_master_bus_init(struct i3c_master_controller *m)
 {
 	struct svc_i3c_master *master = to_svc_i3c_master(m);

@@ -611,6 +661,7 @@ static int svc_i3c_master_bus_init(struct i3c_master_controller *m)
 	      SVC_I3C_MCONFIG_I2CBAUD(i2cbaud);
 	writel(reg, master->regs + SVC_I3C_MCONFIG);
 
+	master->mctrl_config = reg;
 	/* Master core's registration */
 	ret = i3c_master_get_free_addr(m, 0);
 	if (ret < 0)

@@ -1620,6 +1671,7 @@ static const struct i3c_master_controller_ops svc_i3c_master_ops = {
 	.disable_ibi = svc_i3c_master_disable_ibi,
 	.enable_hotjoin = svc_i3c_master_enable_hotjoin,
 	.disable_hotjoin = svc_i3c_master_disable_hotjoin,
+	.set_speed = svc_i3c_master_set_speed,
 };
 
 static int svc_i3c_master_prepare_clks(struct svc_i3c_master *master)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #103 lore raw

PATCH 103/356 i3c: master: svc: Modify enabled_events bit 7:0 to act as IBI enable counter

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Frank Li <[email protected]>

[ Upstream commit 25bc99be5fe53853053ceeaa328068c49dc1e799 ]

Fix issue where disabling IBI on one device disables the entire IBI interrupt. Modify bit 7:0 of enabled_events to serve as an IBI enable counter, ensuring that the system IBI interrupt is disabled only when all I3C devices have IBI disabled.

Fixes 7ff730ca458e ("i3c: master: svc: enable the interrupt in the enable ibi function")

Signed-off-by Frank Li <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Alexandre Belloni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/i3c/master/svc-i3c-master.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

drivers/i3c/master/svc-i3c-master.c +5 -5

index 97d03d755a61f..77bc0db17fc6f 100644
@@ -128,8 +128,8 @@
 /* This parameter depends on the implementation and may be tuned */
 #define SVC_I3C_FIFO_SIZE 16
 
-#define SVC_I3C_EVENT_IBI	BIT(0)
-#define SVC_I3C_EVENT_HOTJOIN	BIT(1)
+#define SVC_I3C_EVENT_IBI	GENMASK(7, 0)
+#define SVC_I3C_EVENT_HOTJOIN	BIT(31)
 
 struct svc_i3c_cmd {
 	u8 addr;

@@ -212,7 +212,7 @@ struct svc_i3c_master {
 		spinlock_t lock;
 	} ibi;
 	struct mutex lock;
-	int enabled_events;
+	u32 enabled_events;
 	u32 mctrl_config;
 };

@@ -1586,7 +1586,7 @@ static int svc_i3c_master_enable_ibi(struct i3c_dev_desc *dev)
 		return ret;
 	}
 
-	master->enabled_events |= SVC_I3C_EVENT_IBI;
+	master->enabled_events++;
 	svc_i3c_master_enable_interrupts(master, SVC_I3C_MINT_SLVSTART);
 
 	return i3c_master_enec_locked(m, dev->info.dyn_addr, I3C_CCC_EVENT_SIR);

@@ -1598,7 +1598,7 @@ static int svc_i3c_master_disable_ibi(struct i3c_dev_desc *dev)
 	struct svc_i3c_master *master = to_svc_i3c_master(m);
 	int ret;
 
-	master->enabled_events &= ~SVC_I3C_EVENT_IBI;
+	master->enabled_events--;
 	if (!master->enabled_events)
 		svc_i3c_master_disable_interrupts(master);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:35 UTC #104 lore raw

PATCH 104/356 i3c: master: Replace hard code 2 with macro I3C_ADDR_SLOT_STATUS_BITS

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Frank Li <[email protected]>

[ Upstream commit 16aed0a6520ba01b7d22c32e193fc1ec674f92d4 ]

Replace the hardcoded value 2, which indicates 2 bits for I3C address status, with the predefined macro I3C_ADDR_SLOT_STATUS_BITS.

Improve maintainability and extensibility of the code.

Signed-off-by Frank Li <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Alexandre Belloni <[email protected]>

Stable-dep-of: 851bd21cdb55 ("i3c: master: Fix dynamic address leak when 'assigned-address' is present")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/i3c/master.c       | 4 ++--
 include/linux/i3c/master.h | 4 +++-
 2 files changed, 5 insertions(+), 3 deletions(-)

drivers/i3c/master.c +2 -2

index 78171a754a3f8..2b6bc03652139 100644
@@ -345,7 +345,7 @@ static enum i3c_addr_slot_status
 i3c_bus_get_addr_slot_status(struct i3c_bus *bus, u16 addr)
 {
 	unsigned long status;
-	int bitpos = addr * 2;
+	int bitpos = addr * I3C_ADDR_SLOT_STATUS_BITS;
 
 	if (addr > I2C_MAX_ADDR)
 		return I3C_ADDR_SLOT_RSVD;

@@ -359,7 +359,7 @@ i3c_bus_get_addr_slot_status(struct i3c_bus *bus, u16 addr)
 static void i3c_bus_set_addr_slot_status(struct i3c_bus *bus, u16 addr,
 					 enum i3c_addr_slot_status status)
 {
-	int bitpos = addr * 2;
+	int bitpos = addr * I3C_ADDR_SLOT_STATUS_BITS;
 	unsigned long *ptr;
 
 	if (addr > I2C_MAX_ADDR)

include/linux/i3c/master.h +3 -1

index f0aa4c549aa7f..ed6b153481021 100644
@@ -305,6 +305,8 @@ enum i3c_addr_slot_status {
 	I3C_ADDR_SLOT_STATUS_MASK = 3,
 };
 
+#define I3C_ADDR_SLOT_STATUS_BITS 2
+
 /**
  * struct i3c_bus - I3C bus object
  * @cur_master: I3C master currently driving the bus. Since I3C is multi-master

@@ -346,7 +348,7 @@ enum i3c_addr_slot_status {
 struct i3c_bus {
 	struct i3c_dev_desc *cur_master;
 	int id;
-	unsigned long addrslots[((I2C_MAX_ADDR + 1) * 2) / BITS_PER_LONG];
+	unsigned long addrslots[((I2C_MAX_ADDR + 1) * I3C_ADDR_SLOT_STATUS_BITS) / BITS_PER_LONG];
 	enum i3c_bus_mode mode;
 	struct {
 		unsigned long i3c;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-12-12 15:36 UTC #105 lore raw

PATCH 105/356 i3c: master: Extend address status bit to 4 and add I3C_ADDR_SLOT_EXT_DESIRED

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Frank Li <[email protected]>

[ Upstream commit 2f552fa280590e61bd3dbe66a7b54b99caa642a4 ]

Extend the address status bit to 4 and introduce the I3C_ADDR_SLOT_EXT_DESIRED macro to indicate that a device prefers a specific address. This is generally set by the 'assigned-address' in the device tree source (dts) file.

 ┌────┬─────────────┬───┬─────────┬───┐
 │S/Sr│ 7'h7E RnW=0 │ACK│ ENTDAA  │ T ├────┐
 └────┴─────────────┴───┴─────────┴───┘    │
 ┌─────────────────────────────────────────┘
 │  ┌──┬─────────────┬───┬─────────────────┬────────────────┬───┬─────────┐
 └─►│Sr│7'h7E RnW=1  │ACK│48bit UID BCR DCR│Assign 7bit Addr│PAR│ ACK/NACK│
    └──┴─────────────┴───┴─────────────────┴────────────────┴───┴─────────┘

Some master controllers (such as HCI) need to prepare the entire above transaction before sending it out to the I3C bus. This means that a 7-bit dynamic address needs to be allocated before knowing the target device's UID information.

However, some I3C targets may request specific addresses (called as "init_dyn_addr"), which is typically specified by the DT-'s assigned-address property. Lower addresses having higher IBI priority. If it is available, i3c_bus_get_free_addr() preferably return a free address that is not in the list of desired addresses (called as "init_dyn_addr"). This allows the device with the "init_dyn_addr" to switch to its "init_dyn_addr" when it hot-joins the I3C bus. Otherwise, if the "init_dyn_addr" is already in use by another I3C device, the target device will not be able to switch to its desired address.

If the previous step fails, fallback returning one of the remaining unassigned address, regardless of its state in the desired list.